Tokenizer: ../arg_m/arg_mining/smlm_pretrained_iter3_0/tokenizer Model: ../arg_m/arg_mining/smlm_pretrained_iter3_0/model
	Train size: 80 Test size: 20


		-------------RUN 1-----------
			------------EPOCH 1---------------
Loss:  tensor(3.6822, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.7724, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4343, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5285, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2417, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7766, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6689, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6303, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6912, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7620, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5667, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5945, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6882, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1916, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5534, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0537, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7542, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7974, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2497, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2802, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5112, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1415, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0998, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1247, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0854, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8642, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0145, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0516, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3270, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8313, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6260, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1857, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9762, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1472, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.08859223300970874, 'recall': 0.15766738660907129, 'f1': 0.11344211344211344, 'number': 463}, 'P': {'precision': 0.4715151515151515, 'recall': 0.5131926121372031, 'f1': 0.4914718888186986, 'number': 758}, 'overall_precision': 0.2801697998787144, 'overall_recall': 0.3783783783783784, 'overall_f1': 0.32195121951219513, 'overall_accuracy': 0.7401525198938992}
			------------EPOCH 2---------------
Loss:  tensor(1.5861, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6443, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2171, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2935, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0985, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6988, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4594, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6414, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4648, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6549, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6190, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6611, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5890, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6396, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4266, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3318, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7531, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2666, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4112, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3963, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5636, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3378, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3421, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2878, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2883, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1758, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1098, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3167, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8280, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4689, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0947, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.22413793103448276, 'recall': 0.4773218142548596, 'f1': 0.30503795721187027, 'number': 463}, 'P': {'precision': 0.5066273932253313, 'recall': 0.45382585751978893, 'f1': 0.47877522616562285, 'number': 758}, 'overall_precision': 0.33933933933933935, 'overall_recall': 0.46273546273546273, 'overall_f1': 0.3915453915453916, 'overall_accuracy': 0.7738395225464191}
			------------EPOCH 3---------------
Loss:  tensor(1.1153, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2863, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1830, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2123, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1339, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2135, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0465, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2149, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3341, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1350, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0794, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3617, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1528, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0077, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1543, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2518, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0708, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7422, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1251, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1558, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0634, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2840, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3971, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0628, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1703, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4260, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0881, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9966, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2904, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9306, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1290, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6866, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0901, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0764, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2920863309352518, 'recall': 0.43844492440604754, 'f1': 0.35060449050086356, 'number': 463}, 'P': {'precision': 0.5508274231678487, 'recall': 0.6147757255936676, 'f1': 0.5810473815461347, 'number': 758}, 'overall_precision': 0.43413367942894227, 'overall_recall': 0.547911547911548, 'overall_f1': 0.4844315713251267, 'overall_accuracy': 0.8130636604774536}
			------------EPOCH 4---------------
Loss:  tensor(0.7489, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9098, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8985, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9458, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0874, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1496, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0387, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0956, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2899, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8376, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8565, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8272, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7508, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7837, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8687, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2734, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0228, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3123, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5407, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2903, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7286, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7105, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7994, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1460, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8255, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8786, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0625, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3496, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7587, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9790, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6745, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2288, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6844, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6755, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7883, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9734, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0590, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4433734939759036, 'recall': 0.39740820734341253, 'f1': 0.41913439635535304, 'number': 463}, 'P': {'precision': 0.5380859375, 'recall': 0.7269129287598944, 'f1': 0.6184062850729518, 'number': 758}, 'overall_precision': 0.5107713690062543, 'overall_recall': 0.601965601965602, 'overall_f1': 0.5526315789473685, 'overall_accuracy': 0.8108090185676392}
			------------EPOCH 5---------------
Loss:  tensor(0.5684, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7107, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6496, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6292, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1265, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0307, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8808, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2382, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7142, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7318, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6889, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5417, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5279, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9687, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2333, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4115, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2353, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5751, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1537, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9966, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5446, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6954, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7674, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2294, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5875, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8173, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5142, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1703, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5255, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2711, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7166, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0266, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4375, 'recall': 0.42332613390928725, 'f1': 0.4302963776070252, 'number': 463}, 'P': {'precision': 0.5423076923076923, 'recall': 0.7440633245382586, 'f1': 0.6273637374860958, 'number': 758}, 'overall_precision': 0.510752688172043, 'overall_recall': 0.6224406224406225, 'overall_f1': 0.5610926541159099, 'overall_accuracy': 0.7949602122015915}
			------------EPOCH 6---------------
Loss:  tensor(0.4841, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5913, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0460, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1166, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0238, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8293, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4737, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4966, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5384, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4505, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3920, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3767, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9395, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4423, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2905, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1885, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3869, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3586, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3696, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8749, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6332, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5137, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1760, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4312, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5964, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2896, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1493, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3625, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2701, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2998, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1739, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6584, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0200, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.371301775147929, 'recall': 0.5421166306695464, 'f1': 0.4407374890254609, 'number': 463}, 'P': {'precision': 0.6192307692307693, 'recall': 0.637203166226913, 'f1': 0.6280884265279584, 'number': 758}, 'overall_precision': 0.5041208791208791, 'overall_recall': 0.6011466011466011, 'overall_f1': 0.5483750466940605, 'overall_accuracy': 0.8270888594164456}
			------------EPOCH 7---------------
Loss:  tensor(0.2811, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2672, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3419, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3529, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0653, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0169, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5191, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1568, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4150, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3489, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4603, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3627, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2420, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2518, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6455, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2490, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1575, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1768, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1337, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3324, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3098, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1341, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7545, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2398, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4571, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3866, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0886, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5854, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2897, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0991, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2584, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2363, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4705, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0088, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.37254901960784315, 'recall': 0.4924406047516199, 'f1': 0.4241860465116279, 'number': 463}, 'P': {'precision': 0.5681818181818182, 'recall': 0.6266490765171504, 'f1': 0.5959849435382685, 'number': 758}, 'overall_precision': 0.48549723756906077, 'overall_recall': 0.5757575757575758, 'overall_f1': 0.5267890595728738, 'overall_accuracy': 0.8183355437665782}
			------------EPOCH 8---------------
Loss:  tensor(0.2065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1540, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2596, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2771, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0191, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0553, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5498, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1482, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4198, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2358, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2208, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1750, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1634, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6879, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3175, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1316, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1377, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0922, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1722, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1998, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0922, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5756, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1768, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3641, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3160, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0848, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2167, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5702, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0632, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1481, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1234, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1149, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0565, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4147, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0058, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4849315068493151, 'recall': 0.38228941684665224, 'f1': 0.4275362318840579, 'number': 463}, 'P': {'precision': 0.578838174273859, 'recall': 0.7361477572559367, 'f1': 0.6480836236933799, 'number': 758}, 'overall_precision': 0.5530474040632054, 'overall_recall': 0.601965601965602, 'overall_f1': 0.5764705882352941, 'overall_accuracy': 0.8075928381962865}
			------------EPOCH 9---------------
Loss:  tensor(0.2196, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2446, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2383, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2196, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0412, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4371, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1265, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2090, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0998, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1379, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1506, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1296, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1345, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5517, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0962, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0541, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0633, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0855, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1779, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0997, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0793, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5781, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1126, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3381, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2176, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0893, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1698, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4547, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1589, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0635, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1655, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0753, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0398, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4361, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0072, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.33633633633633636, 'recall': 0.4838012958963283, 'f1': 0.3968113374667848, 'number': 463}, 'P': {'precision': 0.6174934725848564, 'recall': 0.6240105540897097, 'f1': 0.620734908136483, 'number': 758}, 'overall_precision': 0.486731843575419, 'overall_recall': 0.5708435708435708, 'overall_f1': 0.5254428948360347, 'overall_accuracy': 0.804973474801061}
			------------EPOCH 10---------------
Loss:  tensor(0.1504, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1449, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1132, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0233, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3700, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1665, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1112, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1087, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2415, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0614, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1176, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5327, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0815, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0544, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0830, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0855, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1927, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0935, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0637, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5552, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0725, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3118, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1703, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0464, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0941, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4085, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0965, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0589, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0681, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0603, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0941, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0409, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0052, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.457089552238806, 'recall': 0.5291576673866091, 'f1': 0.49049049049049054, 'number': 463}, 'P': {'precision': 0.6301204819277109, 'recall': 0.6899736147757256, 'f1': 0.6586901763224181, 'number': 758}, 'overall_precision': 0.5622254758418741, 'overall_recall': 0.628992628992629, 'overall_f1': 0.5937379203710863, 'overall_accuracy': 0.825762599469496}
			------------EPOCH 11---------------
Loss:  tensor(0.1154, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0959, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0854, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0861, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0144, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3813, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1201, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1939, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0992, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1165, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1705, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0514, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0338, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0565, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4362, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0555, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0665, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0396, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0703, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0585, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1571, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0864, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0651, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5473, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0478, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2736, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0409, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0712, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4136, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0781, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0665, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0514, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0662, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0585, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0148, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2915, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0043, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3888888888888889, 'recall': 0.5140388768898488, 'f1': 0.4427906976744186, 'number': 463}, 'P': {'precision': 0.6224116930572473, 'recall': 0.6741424802110818, 'f1': 0.6472450918302725, 'number': 758}, 'overall_precision': 0.5226796929518492, 'overall_recall': 0.6134316134316135, 'overall_f1': 0.5644310474755087, 'overall_accuracy': 0.8283488063660478}
			------------EPOCH 12---------------
Loss:  tensor(0.0353, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0360, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0361, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3952, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1303, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1800, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0339, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1523, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0444, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0296, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0437, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4412, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0464, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0475, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0187, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0433, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0485, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0439, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0523, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4135, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2561, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0926, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0321, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0372, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3745, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0371, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0618, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0314, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0339, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0133, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.38646616541353385, 'recall': 0.5550755939524838, 'f1': 0.45567375886524825, 'number': 463}, 'P': {'precision': 0.6175, 'recall': 0.6517150395778364, 'f1': 0.6341463414634146, 'number': 758}, 'overall_precision': 0.5126279863481229, 'overall_recall': 0.6150696150696151, 'overall_f1': 0.5591958302308265, 'overall_accuracy': 0.8292771883289125}
			------------EPOCH 13---------------
Loss:  tensor(0.0501, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0315, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0298, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0328, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2572, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0898, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0890, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0244, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0407, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1105, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0194, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0164, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4421, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0398, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0107, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0387, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0126, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0889, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0312, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0423, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3614, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0102, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1931, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0801, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0242, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0190, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3596, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0506, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0114, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0248, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0164, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1661, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.40834845735027225, 'recall': 0.48596112311015116, 'f1': 0.44378698224852076, 'number': 463}, 'P': {'precision': 0.6036036036036037, 'recall': 0.7071240105540897, 'f1': 0.6512758201701093, 'number': 758}, 'overall_precision': 0.5288394718554552, 'overall_recall': 0.6232596232596233, 'overall_f1': 0.5721804511278196, 'overall_accuracy': 0.8324602122015915}
			------------EPOCH 14---------------
Loss:  tensor(0.0424, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0120, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0150, 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(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1999, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0822, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0698, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0191, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0233, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0920, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0141, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0183, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3790, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0159, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0364, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0340, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0085, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0753, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0247, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0395, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3972, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1683, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0515, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0292, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2876, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0387, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0127, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0214, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0246, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2264, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5170068027210885, 'recall': 0.4924406047516199, 'f1': 0.5044247787610621, 'number': 463}, 'P': {'precision': 0.6058315334773218, 'recall': 0.7401055408970977, 'f1': 0.666270783847981, 'number': 758}, 'overall_precision': 0.5771762984637894, 'overall_recall': 0.6461916461916462, 'overall_f1': 0.6097372488408037, 'overall_accuracy': 0.8242042440318302}
			------------EPOCH 15---------------
Loss:  tensor(0.0138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0150, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0083, 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(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3251, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0756, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0137, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0234, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0817, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0139, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3666, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0149, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0318, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0262, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0105, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0711, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0570, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1891, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0129, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3954, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0135, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0463, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0184, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1488, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4989384288747346, 'recall': 0.5075593952483801, 'f1': 0.5032119914346895, 'number': 463}, 'P': {'precision': 0.629802095459837, 'recall': 0.7137203166226913, 'f1': 0.6691403834260977, 'number': 758}, 'overall_precision': 0.5834586466165413, 'overall_recall': 0.6355446355446356, 'overall_f1': 0.6083888671109369, 'overall_accuracy': 0.8244363395225465}
			------------EPOCH 16---------------
Loss:  tensor(0.0202, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0198, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0133, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2258, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0767, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0638, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0118, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0867, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0114, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0127, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3644, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0240, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0114, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0247, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0319, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3527, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1498, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0227, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3181, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1168, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4605263157894737, 'recall': 0.5291576673866091, 'f1': 0.4924623115577889, 'number': 463}, 'P': {'precision': 0.6917197452229299, 'recall': 0.716358839050132, 'f1': 0.7038237200259235, 'number': 758}, 'overall_precision': 0.59832953682612, 'overall_recall': 0.6453726453726454, 'overall_f1': 0.6209613869188338, 'overall_accuracy': 0.8337864721485411}
			------------EPOCH 17---------------
Loss:  tensor(0.0187, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0130, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0254, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0142, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1962, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0677, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0456, 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(0.0582, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0228, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0513, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0284, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0373, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2936, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0174, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0090, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0368, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0201, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4585537918871252, 'recall': 0.5615550755939525, 'f1': 0.5048543689320388, 'number': 463}, 'P': {'precision': 0.6696891191709845, 'recall': 0.6820580474934037, 'f1': 0.6758169934640524, 'number': 758}, 'overall_precision': 0.5802837938760269, 'overall_recall': 0.6363636363636364, 'overall_f1': 0.60703125, 'overall_accuracy': 0.8291777188328913}
			------------EPOCH 18---------------
Loss:  tensor(0.0036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0799, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0565, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0136, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0585, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0102, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2929, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0184, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0588, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0366, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3450, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1685, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0118, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2320, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0382, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0038, 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(0.1146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.444636678200692, 'recall': 0.5550755939524838, 'f1': 0.49375600384245916, 'number': 463}, 'P': {'precision': 0.6466992665036675, 'recall': 0.6978891820580475, 'f1': 0.6713197969543148, 'number': 758}, 'overall_precision': 0.5630372492836676, 'overall_recall': 0.6437346437346437, 'overall_f1': 0.6006878104700037, 'overall_accuracy': 0.8309681697612732}
			------------EPOCH 19---------------
Loss:  tensor(0.0058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0715, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0697, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0896, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3292, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0140, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0437, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0320, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2977, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1325, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0185, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0482, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0127, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2148, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.44212523719165087, 'recall': 0.5032397408207343, 'f1': 0.4707070707070707, 'number': 463}, 'P': {'precision': 0.6030871003307607, 'recall': 0.7216358839050132, 'f1': 0.6570570570570571, 'number': 758}, 'overall_precision': 0.5439330543933054, 'overall_recall': 0.6388206388206388, 'overall_f1': 0.5875706214689266, 'overall_accuracy': 0.8253978779840849}
			------------EPOCH 20---------------
Loss:  tensor(0.0038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0046, 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(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1912, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0776, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0439, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0211, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0653, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0148, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2730, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0142, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0323, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2876, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1370, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0160, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0174, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2164, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0390, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0355, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0135, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.48125, 'recall': 0.49892008639308855, 'f1': 0.4899257688229056, 'number': 463}, 'P': {'precision': 0.6043715846994535, 'recall': 0.7295514511873351, 'f1': 0.6610878661087866, 'number': 758}, 'overall_precision': 0.5620071684587814, 'overall_recall': 0.6420966420966421, 'overall_f1': 0.5993883792048929, 'overall_accuracy': 0.8261936339522546}
			------------EPOCH 21---------------
Loss:  tensor(0.0029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, 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(0.2060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0560, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0371, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0167, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0562, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0254, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0155, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2626, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0127, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0262, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2276, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0332, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0163, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0185, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0647, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0432, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0719, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4191304347826087, 'recall': 0.5205183585313174, 'f1': 0.4643545279383429, 'number': 463}, 'P': {'precision': 0.6063829787234043, 'recall': 0.6767810026385225, 'f1': 0.6396508728179551, 'number': 758}, 'overall_precision': 0.5306122448979592, 'overall_recall': 0.6175266175266175, 'overall_f1': 0.570779712339137, 'overall_accuracy': 0.8233090185676393}
			------------EPOCH 22---------------
Loss:  tensor(0.0042, 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(0.0036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1251, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0213, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0169, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0322, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2290, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0132, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0299, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0258, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0292, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2436, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0121, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0214, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1586, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0302, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0118, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0156, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0495, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.41821561338289964, 'recall': 0.48596112311015116, 'f1': 0.4495504495504495, 'number': 463}, 'P': {'precision': 0.619277108433735, 'recall': 0.6781002638522428, 'f1': 0.6473551637279598, 'number': 758}, 'overall_precision': 0.5402046783625731, 'overall_recall': 0.6052416052416052, 'overall_f1': 0.570876786404017, 'overall_accuracy': 0.8280172413793103}
			------------EPOCH 23---------------
Loss:  tensor(0.0130, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0120, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0169, 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(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1517, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0549, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0295, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0155, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0303, 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(0.0075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2616, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0280, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0348, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0929, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0156, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1652, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0407, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0241, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0536, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4081237911025145, 'recall': 0.4557235421166307, 'f1': 0.4306122448979592, 'number': 463}, 'P': {'precision': 0.5624332977588047, 'recall': 0.6952506596306068, 'f1': 0.6218289085545723, 'number': 758}, 'overall_precision': 0.5075653370013755, 'overall_recall': 0.6044226044226044, 'overall_f1': 0.5517757009345795, 'overall_accuracy': 0.8238395225464191}
			------------EPOCH 24---------------
Loss:  tensor(0.0066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0075, 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(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1449, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0364, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0258, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0159, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0316, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2086, 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(0.0047, 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(0.0364, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0111, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0426, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0900, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0227, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1506, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0098, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0325, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0120, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0601, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5663265306122449, 'recall': 0.4794816414686825, 'f1': 0.5192982456140351, 'number': 463}, 'P': {'precision': 0.5269749518304432, 'recall': 0.7216358839050132, 'f1': 0.6091314031180401, 'number': 758}, 'overall_precision': 0.5377622377622377, 'overall_recall': 0.6298116298116299, 'overall_f1': 0.5801584307808375, 'overall_accuracy': 0.8249336870026526}
			------------EPOCH 25---------------
Loss:  tensor(0.0088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1695, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0572, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0385, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0118, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0314, 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(0.0092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1967, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0084, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0020, 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(0.0195, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0240, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1919, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0629, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0333, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1535, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0380, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0027, 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(0.0579, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0001, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4652908067542214, 'recall': 0.5356371490280778, 'f1': 0.497991967871486, 'number': 463}, 'P': {'precision': 0.6507936507936508, 'recall': 0.7031662269129287, 'f1': 0.6759670259987318, 'number': 758}, 'overall_precision': 0.5776627218934911, 'overall_recall': 0.6396396396396397, 'overall_f1': 0.6070734551107657, 'overall_accuracy': 0.8277851458885942}
			------------EPOCH 26---------------
Loss:  tensor(0.0060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, 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(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1123, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0251, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0136, 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(0.0203, 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(0.0050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1738, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0141, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0355, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0614, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1359, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0308, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0087, 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(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0363, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4366438356164384, 'recall': 0.550755939524838, 'f1': 0.4871060171919771, 'number': 463}, 'P': {'precision': 0.6223862238622386, 'recall': 0.6675461741424802, 'f1': 0.6441756842775302, 'number': 758}, 'overall_precision': 0.5447387258410881, 'overall_recall': 0.6232596232596233, 'overall_f1': 0.5813598166539343, 'overall_accuracy': 0.8284814323607427}
			------------EPOCH 27---------------
Loss:  tensor(0.0063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0015, 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(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0962, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0327, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0261, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0307, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2346, 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(0.0011, 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(0.0123, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1658, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0580, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0063, 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(0.1232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0393, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0076, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0430, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0001, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5136363636363637, 'recall': 0.48812095032397407, 'f1': 0.5005537098560354, 'number': 463}, 'P': {'precision': 0.6223929747530187, 'recall': 0.7480211081794196, 'f1': 0.6794487717195926, 'number': 758}, 'overall_precision': 0.5869726128793487, 'overall_recall': 0.6494676494676495, 'overall_f1': 0.6166407465007776, 'overall_accuracy': 0.8291114058355438}
			------------EPOCH 28---------------
Loss:  tensor(0.0061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, 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(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.7575e-05, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0912, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0256, 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(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0190, 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(0.0042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1476, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0024, 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(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0231, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1467, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0435, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0952, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0291, 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(0.0080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0183, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0001, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4819532908704883, 'recall': 0.490280777537797, 'f1': 0.4860813704496788, 'number': 463}, 'P': {'precision': 0.6205714285714286, 'recall': 0.716358839050132, 'f1': 0.6650336803429271, 'number': 758}, 'overall_precision': 0.5720653789004457, 'overall_recall': 0.6306306306306306, 'overall_f1': 0.5999220880405142, 'overall_accuracy': 0.8280503978779841}
			------------EPOCH 29---------------
Loss:  tensor(0.0051, 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(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0226e-05, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0482, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0173, 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(0.0135, 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(0.0037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1312, 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(0.0008, 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(0.0061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0177, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1390, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0353, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0763, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0240, 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(0.0086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0242, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0320e-05, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.44368600682593856, 'recall': 0.5615550755939525, 'f1': 0.4957102001906578, 'number': 463}, 'P': {'precision': 0.6224116930572473, 'recall': 0.6741424802110818, 'f1': 0.6472450918302725, 'number': 758}, 'overall_precision': 0.5479744136460555, 'overall_recall': 0.6314496314496314, 'overall_f1': 0.58675799086758, 'overall_accuracy': 0.8281498673740053}
			------------EPOCH 30---------------
Loss:  tensor(0.0036, 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(0.0014, 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(0.0001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0660, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0140, 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(0.0133, 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(0.0037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1091, 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(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0098, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0154, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1287, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0746, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0240, 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(0.0065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0251, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.5779e-05, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4854771784232365, 'recall': 0.5053995680345572, 'f1': 0.49523809523809526, 'number': 463}, 'P': {'precision': 0.6178489702517163, 'recall': 0.712401055408971, 'f1': 0.6617647058823529, 'number': 758}, 'overall_precision': 0.5707964601769911, 'overall_recall': 0.6339066339066339, 'overall_f1': 0.6006984866123399, 'overall_accuracy': 0.8274535809018567}


		-------------RUN 2-----------
			------------EPOCH 1---------------
Loss:  tensor(0.5097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5952, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4702, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4298, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4406, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2508, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2888, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7515, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9430, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8340, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8546, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8374, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6952, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5781, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1578, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4705, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2884, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4886, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3691, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6299, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4276, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5301, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1910, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7268, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0581, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0521, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1531, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0332, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2756, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0165, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8459, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8923, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7415, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6941, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8870, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.17616580310880828, 'recall': 0.278118609406953, 'f1': 0.21570182394924664, 'number': 489}, 'P': {'precision': 0.3175, 'recall': 0.42474916387959866, 'f1': 0.36337625178826893, 'number': 897}, 'overall_precision': 0.26217038539553755, 'overall_recall': 0.373015873015873, 'overall_f1': 0.307921381774866, 'overall_accuracy': 0.7123566513761468}
			------------EPOCH 2---------------
Loss:  tensor(0.2664, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3190, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8516, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0904, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4169, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7922, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7309, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1584, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1978, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5262, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5550, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8392, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5327, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1731, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5210, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4491, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4925, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6468, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9602, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0150, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0412, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4105, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4911, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4418, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4593, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3513, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1749, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4910, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2565, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4150, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2383, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3416, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3179190751445087, 'recall': 0.5623721881390593, 'f1': 0.40620384047267355, 'number': 489}, 'P': {'precision': 0.5200433369447454, 'recall': 0.5351170568561873, 'f1': 0.5274725274725275, 'number': 897}, 'overall_precision': 0.4222595078299776, 'overall_recall': 0.5447330447330447, 'overall_f1': 0.47574039067422813, 'overall_accuracy': 0.7643635321100918}
			------------EPOCH 3---------------
Loss:  tensor(0.1763, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2213, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2261, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0559, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2968, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5316, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5615, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0923, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7868, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4182, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0774, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2147, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8895, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0964, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8387, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0567, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3473, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1254, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0401, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3818, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3297, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7352, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8563, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7483, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8575, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3897, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0844, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0547, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0133, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0984, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1362, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9858, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1470, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9889, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9584, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0823, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.45121951219512196, 'recall': 0.5296523517382413, 'f1': 0.4873000940733772, 'number': 489}, 'P': {'precision': 0.6045627376425855, 'recall': 0.7090301003344481, 'f1': 0.6526423807080554, 'number': 897}, 'overall_precision': 0.550430504305043, 'overall_recall': 0.6457431457431457, 'overall_f1': 0.594289508632138, 'overall_accuracy': 0.8077121559633027}
			------------EPOCH 4---------------
Loss:  tensor(0.1485, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1985, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0275, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3115, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3633, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0802, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5439, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2778, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8451, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9661, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9358, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7240, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7770, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6120, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7500, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8803, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0129, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7411, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1830, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1871, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6357, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7693, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5955, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7924, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2975, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7835, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8143, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7331, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9410, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0690, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9381, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8484, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9612, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6296, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8649, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.34836852207293667, 'recall': 0.7423312883435583, 'f1': 0.47419986936642716, 'number': 489}, 'P': {'precision': 0.696078431372549, 'recall': 0.47491638795986624, 'f1': 0.5646123260437376, 'number': 897}, 'overall_precision': 0.47702539298669894, 'overall_recall': 0.5692640692640693, 'overall_f1': 0.5190789473684211, 'overall_accuracy': 0.7354071100917431}
			------------EPOCH 5---------------
Loss:  tensor(0.1054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1287, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8718, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0912, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3493, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4236, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0681, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6805, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3312, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7150, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8811, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7857, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5320, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5735, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6487, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8338, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8504, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7246, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6418, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0274, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9105, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4612, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5751, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3243, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3795, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4920, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2302, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5632, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5728, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5467, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5404, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0540, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6477, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6114, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7687, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4972, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4972, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6211, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.39858156028368796, 'recall': 0.5746421267893661, 'f1': 0.4706867671691793, 'number': 489}, 'P': {'precision': 0.6305015353121801, 'recall': 0.6867335562987736, 'f1': 0.6574172892209178, 'number': 897}, 'overall_precision': 0.5332936979785969, 'overall_recall': 0.6471861471861472, 'overall_f1': 0.5847457627118645, 'overall_accuracy': 0.8117259174311927}
			------------EPOCH 6---------------
Loss:  tensor(0.1012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5660, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0249, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2243, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2945, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0240, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1975, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4581, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5967, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5695, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3912, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4364, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3490, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3746, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4704, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6530, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3706, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7941, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7312, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3237, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4605, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2098, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3299, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3969, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1406, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4296, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3414, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4180, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0351, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4712, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3313, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2500, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4179, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3857331571994716, 'recall': 0.5971370143149284, 'f1': 0.46869983948635635, 'number': 489}, 'P': {'precision': 0.6543075245365322, 'recall': 0.6688963210702341, 'f1': 0.6615214994487321, 'number': 897}, 'overall_precision': 0.5328554360812425, 'overall_recall': 0.6435786435786436, 'overall_f1': 0.5830065359477125, 'overall_accuracy': 0.8040424311926605}
			------------EPOCH 7---------------
Loss:  tensor(0.0742, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3390, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2218, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1941, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0213, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2851, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1670, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2584, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2485, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3283, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2923, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3952, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5927, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2499, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2549, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6794, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7335, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2605, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4809, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2276, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2568, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2318, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2724, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3239, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3850, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0527, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6905, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5130, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4958, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1916, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2946, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5657492354740061, 'recall': 0.3783231083844581, 'f1': 0.45343137254901966, 'number': 489}, 'P': {'precision': 0.48469750889679714, 'recall': 0.7591973244147158, 'f1': 0.5916594265855778, 'number': 897}, 'overall_precision': 0.5, 'overall_recall': 0.6248196248196248, 'overall_f1': 0.5554842847979473, 'overall_accuracy': 0.8029529816513762}
			------------EPOCH 8---------------
Loss:  tensor(0.1267, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1234, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7567, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0342, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3427, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0285, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1873, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1854, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2213, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3782, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4175, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2364, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3180, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1918, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2704, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2652, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6930, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2695, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3254, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7234, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5465, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2292, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3929, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1872, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2607, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4998, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3850, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3369, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0121, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3705, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2182, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2194, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2982, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6374, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3232421875, 'recall': 0.6768916155419223, 'f1': 0.4375413086582948, 'number': 489}, 'P': {'precision': 0.4975669099756691, 'recall': 0.4559643255295429, 'f1': 0.4758580570098894, 'number': 897}, 'overall_precision': 0.4008667388949079, 'overall_recall': 0.5339105339105339, 'overall_f1': 0.457920792079208, 'overall_accuracy': 0.7659977064220184}
			------------EPOCH 9---------------
Loss:  tensor(0.0589, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1466, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5167, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0439, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1639, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1561, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0238, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2863, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3000, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2674, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2436, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5153, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1289, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2236, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4274, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1680, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3195, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0912, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2763, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1728, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1870, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2311, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2509, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3116, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2187, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1302, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2177, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3635280095351609, 'recall': 0.623721881390593, 'f1': 0.45933734939759036, 'number': 489}, 'P': {'precision': 0.6832740213523132, 'recall': 0.6421404682274248, 'f1': 0.6620689655172414, 'number': 897}, 'overall_precision': 0.523781212841855, 'overall_recall': 0.6356421356421357, 'overall_f1': 0.5743155149934811, 'overall_accuracy': 0.807454128440367}
			------------EPOCH 10---------------
Loss:  tensor(0.0106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0268, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1624, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0801, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0612, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1286, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1973, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2780, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2566, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1668, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2527, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1667, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1689, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1608, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4493, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1149, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1936, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3633, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2424, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0526, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0739, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1136, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0422, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0771, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0731, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0953, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1956, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2166, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4421, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0966, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2050, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4349112426035503, 'recall': 0.6012269938650306, 'f1': 0.5047210300429185, 'number': 489}, 'P': {'precision': 0.7029077117572693, 'recall': 0.6198439241917503, 'f1': 0.6587677725118485, 'number': 897}, 'overall_precision': 0.5794137695978187, 'overall_recall': 0.6132756132756133, 'overall_f1': 0.5958640028040659, 'overall_accuracy': 0.7936353211009174}
			------------EPOCH 11---------------
Loss:  tensor(0.0113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0201, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1322, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0180, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0965, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0872, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0313, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1823, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2540, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0937, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2821, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2692, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1266, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1182, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0666, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0665, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3848, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0536, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1508, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4380, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2976, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0861, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2424, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0371, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0737, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0600, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0268, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0874, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2213, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2275, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0857, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3156, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0407, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0460, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1369, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.39389920424403185, 'recall': 0.6073619631901841, 'f1': 0.47787610619469023, 'number': 489}, 'P': {'precision': 0.7032828282828283, 'recall': 0.620958751393534, 'f1': 0.6595618709295441, 'number': 897}, 'overall_precision': 0.5523932729624839, 'overall_recall': 0.6161616161616161, 'overall_f1': 0.582537517053206, 'overall_accuracy': 0.8091743119266055}
			------------EPOCH 12---------------
Loss:  tensor(0.0037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0848, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0665, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0611, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0378, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2478, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1561, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0809, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2986, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1877, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0978, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1174, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1145, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0594, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0761, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3575, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0368, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1187, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4370, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2384, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0527, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2170, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0234, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0436, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0364, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0132, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0566, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0541, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0548, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1934, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4437, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0623, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1352, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4485981308411215, 'recall': 0.588957055214724, 'f1': 0.5092838196286472, 'number': 489}, 'P': {'precision': 0.5828460038986355, 'recall': 0.6666666666666666, 'f1': 0.6219448777951118, 'number': 897}, 'overall_precision': 0.5311750599520384, 'overall_recall': 0.6392496392496393, 'overall_f1': 0.5802226588081206, 'overall_accuracy': 0.8257740825688074}
			------------EPOCH 13---------------
Loss:  tensor(0.0030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0076, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0497, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0475, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0413, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0159, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1657, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1279, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0387, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1476, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1513, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0701, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0875, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1126, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0612, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0763, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0695, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1741, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4608, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2218, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0450, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0234, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0292, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0259, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0312, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1696, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0301, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3210, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0333, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0480, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0959, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5514705882352942, 'recall': 0.4601226993865031, 'f1': 0.5016722408026756, 'number': 489}, 'P': {'precision': 0.46187683284457476, 'recall': 0.7023411371237458, 'f1': 0.5572755417956655, 'number': 897}, 'overall_precision': 0.48250564334085777, 'overall_recall': 0.6168831168831169, 'overall_f1': 0.5414819506016466, 'overall_accuracy': 0.8189793577981651}
			------------EPOCH 14---------------
Loss:  tensor(0.0073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0247, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0754, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0605, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0293, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1283, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1899, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0551, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1759, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1856, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0509, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0475, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0868, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0512, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0443, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3142, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0352, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0931, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3944, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1835, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0464, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2163, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0128, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0445, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0163, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0288, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0717, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0780, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1199, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2614, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0387, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0154, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0781, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.35931307793923384, 'recall': 0.556237218813906, 'f1': 0.43659711075441415, 'number': 489}, 'P': {'precision': 0.6522727272727272, 'recall': 0.6399108138238573, 'f1': 0.6460326392796849, 'number': 897}, 'overall_precision': 0.5167990226023214, 'overall_recall': 0.6103896103896104, 'overall_f1': 0.5597088984452532, 'overall_accuracy': 0.8148509174311926}
			------------EPOCH 15---------------
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0478, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0596, 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(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0479, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1298, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0973, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0240, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0211, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0813, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0481, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0433, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2652, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0320, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0792, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3400, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1598, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0252, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1766, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0227, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0147, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0141, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0129, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0852, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0674, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2637, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1989, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0218, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0085, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0292, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.39502762430939226, 'recall': 0.5848670756646217, 'f1': 0.47155812036273703, 'number': 489}, 'P': {'precision': 0.6260504201680672, 'recall': 0.6644370122630993, 'f1': 0.6446727961060033, 'number': 897}, 'overall_precision': 0.5262529832935561, 'overall_recall': 0.6363636363636364, 'overall_f1': 0.5760940561724363, 'overall_accuracy': 0.8195240825688074}
			------------EPOCH 16---------------
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0212, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0368, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0240, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0993, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0176, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0827, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0945, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0199, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0133, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0632, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0286, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0155, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0577, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3320, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1142, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0163, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1554, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0185, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0237, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0851, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1749, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0285, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2694, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0224, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4379905808477237, 'recall': 0.5705521472392638, 'f1': 0.4955595026642985, 'number': 489}, 'P': {'precision': 0.6394984326018809, 'recall': 0.6822742474916388, 'f1': 0.6601941747572815, 'number': 897}, 'overall_precision': 0.5589711417816813, 'overall_recall': 0.6428571428571429, 'overall_f1': 0.5979865771812081, 'overall_accuracy': 0.8170584862385321}
			------------EPOCH 17---------------
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0147, 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>)
Loss:  tensor(0.0317, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0614, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0994, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0975, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0466, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0268, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0417, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0319, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2433, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0171, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0567, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1180, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0186, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1471, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0027, 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(0.0050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0105, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0152, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0615, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1786, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0313, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1523, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0140, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0320, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.43769470404984423, 'recall': 0.5746421267893661, 'f1': 0.49690539345711765, 'number': 489}, 'P': {'precision': 0.6722037652270211, 'recall': 0.6767001114827201, 'f1': 0.6744444444444445, 'number': 897}, 'overall_precision': 0.574757281553398, 'overall_recall': 0.6406926406926406, 'overall_f1': 0.6059365404298874, 'overall_accuracy': 0.8181192660550459}
			------------EPOCH 18---------------
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0297, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0182, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0406, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0607, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0273, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0607, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0313, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2407, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0451, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2765, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0868, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0129, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1227, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0123, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0359, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0107, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0141, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0195, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0645, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1596, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0186, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1807, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0166, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0258, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0559, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5119047619047619, 'recall': 0.5276073619631901, 'f1': 0.5196374622356494, 'number': 489}, 'P': {'precision': 0.614100185528757, 'recall': 0.738015607580825, 'f1': 0.670379746835443, 'number': 897}, 'overall_precision': 0.5815423514538559, 'overall_recall': 0.6637806637806638, 'overall_f1': 0.6199460916442048, 'overall_accuracy': 0.8294724770642202}
			------------EPOCH 19---------------
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0306, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0260, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0422, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1207, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0563, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0489, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2234, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0188, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0435, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2658, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0760, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0188, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0341, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0348, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1269, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0126, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1213, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0224, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5384615384615384, 'recall': 0.5296523517382413, 'f1': 0.534020618556701, 'number': 489}, 'P': {'precision': 0.6539618856569709, 'recall': 0.7268673355629878, 'f1': 0.6884899683210137, 'number': 897}, 'overall_precision': 0.6163734776725305, 'overall_recall': 0.6572871572871573, 'overall_f1': 0.6361731843575419, 'overall_accuracy': 0.8224483944954128}
			------------EPOCH 20---------------
Loss:  tensor(0.0056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0303, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0584, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0693, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1597, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0247, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1379, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0684, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0486, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0203, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1975, 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(0.2288, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0770, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0126, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1283, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0436, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0625, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0181, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0249, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0190, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0634, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1236, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1823, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0227, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0235, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4533762057877814, 'recall': 0.5766871165644172, 'f1': 0.5076507650765077, 'number': 489}, 'P': {'precision': 0.6805251641137856, 'recall': 0.693422519509476, 'f1': 0.6869133075648813, 'number': 897}, 'overall_precision': 0.5885416666666666, 'overall_recall': 0.6522366522366523, 'overall_f1': 0.6187542778918548, 'overall_accuracy': 0.8210435779816514}
			------------EPOCH 21---------------
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0321, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0355, 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(0.0431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0999, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0171, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0412, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0652, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0346, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0633, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0514, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0087, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0401, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1971, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0334, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2412, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0775, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0098, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0077, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0207, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0193, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0680, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1486, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0445, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1675, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0483, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0182, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0334, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4155455904334828, 'recall': 0.5685071574642127, 'f1': 0.4801381692573402, 'number': 489}, 'P': {'precision': 0.6524973432518597, 'recall': 0.6845039018952063, 'f1': 0.6681175190424373, 'number': 897}, 'overall_precision': 0.5540372670807453, 'overall_recall': 0.6435786435786436, 'overall_f1': 0.5954606141522029, 'overall_accuracy': 0.8213876146788991}
			------------EPOCH 22---------------
Loss:  tensor(0.0005, 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(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0302, 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(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0883, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0321, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0132, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0352, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0437, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1754, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0282, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0381, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2303, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0708, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0208, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0287, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1487, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0929, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0107, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4467408585055644, 'recall': 0.5746421267893661, 'f1': 0.5026833631484794, 'number': 489}, 'P': {'precision': 0.53584229390681, 'recall': 0.6666666666666666, 'f1': 0.5941381023348237, 'number': 897}, 'overall_precision': 0.5037249283667622, 'overall_recall': 0.6341991341991342, 'overall_f1': 0.5614819546470776, 'overall_accuracy': 0.8232511467889908}
			------------EPOCH 23---------------
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0148, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0288, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0196, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0653, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0396, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0462, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0327, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0654, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0203, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1672, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0292, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0247, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2182, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0692, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0248, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0906, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0054, 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(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0252, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1137, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0774, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0066, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.45733788395904434, 'recall': 0.5480572597137015, 'f1': 0.4986046511627908, 'number': 489}, 'P': {'precision': 0.6583592938733126, 'recall': 0.7068004459308808, 'f1': 0.681720430107527, 'number': 897}, 'overall_precision': 0.5823111684958038, 'overall_recall': 0.6507936507936508, 'overall_f1': 0.6146507666098808, 'overall_accuracy': 0.8171731651376147}
			------------EPOCH 24---------------
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0230, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0134, 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(0.0461, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0700, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0316, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0303, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0367, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1422, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0136, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1759, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0554, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0840, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0084, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0218, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1316, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0827, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0059, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.48854961832061067, 'recall': 0.523517382413088, 'f1': 0.5054294175715696, 'number': 489}, 'P': {'precision': 0.6475814412635735, 'recall': 0.7313266443701226, 'f1': 0.686910994764398, 'number': 897}, 'overall_precision': 0.5933636955107352, 'overall_recall': 0.658008658008658, 'overall_f1': 0.624016421484776, 'overall_accuracy': 0.8216743119266054}
			------------EPOCH 25---------------
Loss:  tensor(0.0004, 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(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0180, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0467, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0614, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0166, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0297, 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(0.0368, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0087, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1309, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0133, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1661, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0758, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0028, 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(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0171, 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(0.0033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0646, 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(0.0069, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4251207729468599, 'recall': 0.5398773006134969, 'f1': 0.4756756756756757, 'number': 489}, 'P': {'precision': 0.6369168356997972, 'recall': 0.7001114827201784, 'f1': 0.667020711630377, 'number': 897}, 'overall_precision': 0.5550715619166148, 'overall_recall': 0.6435786435786436, 'overall_f1': 0.5960574674239892, 'overall_accuracy': 0.8188933486238532}
			------------EPOCH 26---------------
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0190, 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(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0496, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0529, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0155, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0257, 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(0.0441, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1190, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0098, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1478, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0318, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0530, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0036, 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>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1441, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0594, 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(0.0293, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4907749077490775, 'recall': 0.5439672801635992, 'f1': 0.516003879728419, 'number': 489}, 'P': {'precision': 0.6498993963782697, 'recall': 0.7201783723522854, 'f1': 0.6832363828662085, 'number': 897}, 'overall_precision': 0.59375, 'overall_recall': 0.658008658008658, 'overall_f1': 0.6242299794661191, 'overall_accuracy': 0.8212729357798165}
			------------EPOCH 27---------------
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0173, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0045, 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(0.0387, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0486, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0129, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0208, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, 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(0.0059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1391, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0244, 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(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0029, 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>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0962, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0414, 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(0.0066, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4240506329113924, 'recall': 0.5480572597137015, 'f1': 0.47814451382694023, 'number': 489}, 'P': {'precision': 0.6286579212916246, 'recall': 0.6945373467112598, 'f1': 0.659957627118644, 'number': 897}, 'overall_precision': 0.5489833641404805, 'overall_recall': 0.6428571428571429, 'overall_f1': 0.5922233300099701, 'overall_accuracy': 0.8184633027522936}
			------------EPOCH 28---------------
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.4305e-05, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0026, 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(0.0495, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0499, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0187, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1207, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1460, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0261, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0483, 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(0.0024, 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>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0140, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0412, 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(0.0045, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4849056603773585, 'recall': 0.5255623721881391, 'f1': 0.5044160942100099, 'number': 489}, 'P': {'precision': 0.6309639727361246, 'recall': 0.7224080267558528, 'f1': 0.6735966735966736, 'number': 897}, 'overall_precision': 0.5812459858702633, 'overall_recall': 0.652958152958153, 'overall_f1': 0.6150186884131837, 'overall_accuracy': 0.8221043577981652}
			------------EPOCH 29---------------
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.8788e-05, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0193, 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(0.0001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0445, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0154, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0209, 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(0.0269, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0085, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1400, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0178, 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(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, 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(0.0043, 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(0.1004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0237, 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(0.0038, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4401294498381877, 'recall': 0.556237218813906, 'f1': 0.4914182475158085, 'number': 489}, 'P': {'precision': 0.6231155778894473, 'recall': 0.6911928651059086, 'f1': 0.6553911205073996, 'number': 897}, 'overall_precision': 0.5530068195908245, 'overall_recall': 0.6435786435786436, 'overall_f1': 0.594864954984995, 'overall_accuracy': 0.8187786697247706}
			------------EPOCH 30---------------
Loss:  tensor(0.0002, 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(7.6784e-05, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0164, 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(0.0001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0489, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0425, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0110, 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(0.0292, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0944, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0090, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1273, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0171, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0334, 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(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, 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(0.0043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1144, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0032, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4517684887459807, 'recall': 0.5746421267893661, 'f1': 0.5058505850585059, 'number': 489}, 'P': {'precision': 0.6623931623931624, 'recall': 0.6911928651059086, 'f1': 0.6764866339334424, 'number': 897}, 'overall_precision': 0.5783055198973043, 'overall_recall': 0.65007215007215, 'overall_f1': 0.6120923913043479, 'overall_accuracy': 0.8171158256880734}


		-------------RUN 3-----------
			------------EPOCH 1---------------
Loss:  tensor(3.6075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1863, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2996, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0542, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0366, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4107, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5571, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7535, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8591, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2903, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3887, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4930, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5716, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1919, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2642, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3718, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2868, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3170, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1577, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3849, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3746, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1466, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1846, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2154, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7341, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4870, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7781, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3642, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8383, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8659, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2722, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6542, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6874, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2327, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.25654450261780104, 'recall': 0.11448598130841121, 'f1': 0.1583198707592892, 'number': 428}, 'P': {'precision': 0.4287200832466181, 'recall': 0.5988372093023255, 'f1': 0.49969678593086714, 'number': 688}, 'overall_precision': 0.4001736111111111, 'overall_recall': 0.41308243727598565, 'overall_f1': 0.40652557319223986, 'overall_accuracy': 0.737195663763995}
			------------EPOCH 2---------------
Loss:  tensor(1.6986, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5808, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5343, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5437, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6105, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6821, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6307, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5587, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6285, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0893, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6828, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3402, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4864, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0815, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2795, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4200, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5758, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4683, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2839, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2163, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8535, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1957, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7825, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1238, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1220, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4803, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8382, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0590, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1422, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2750, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9370, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9869, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8043, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.42780748663101603, 'recall': 0.5607476635514018, 'f1': 0.4853387259858442, 'number': 428}, 'P': {'precision': 0.5233516483516484, 'recall': 0.5537790697674418, 'f1': 0.538135593220339, 'number': 688}, 'overall_precision': 0.48176881303335917, 'overall_recall': 0.5564516129032258, 'overall_f1': 0.5164241164241163, 'overall_accuracy': 0.8131508796872223}
			------------EPOCH 3---------------
Loss:  tensor(1.2305, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1999, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9678, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0931, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9675, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1586, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0511, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0913, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9653, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7102, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2502, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1696, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9711, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9911, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2262, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0467, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8555, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0715, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0549, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8668, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7830, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1482, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8343, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9159, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5577, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7916, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2816, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8933, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3484, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5617, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6813, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7880, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2409, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6458, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7715, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6295, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4743362831858407, 'recall': 0.6261682242990654, 'f1': 0.5397784491440081, 'number': 428}, 'P': {'precision': 0.603641456582633, 'recall': 0.626453488372093, 'f1': 0.6148359486447931, 'number': 688}, 'overall_precision': 0.5465207193119624, 'overall_recall': 0.6263440860215054, 'overall_f1': 0.5837160751565762, 'overall_accuracy': 0.834441087613293}
			------------EPOCH 4---------------
Loss:  tensor(0.9100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6488, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8193, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6633, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2962, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6286, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9534, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6537, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9586, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0529, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8112, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9428, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7731, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6276, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6251, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4404, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7535, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7384, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4669, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5582, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2350, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2785, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7139, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6796, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2135, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4991, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5482, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5251, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5429553264604811, 'recall': 0.3691588785046729, 'f1': 0.4394993045897079, 'number': 428}, 'P': {'precision': 0.5390070921985816, 'recall': 0.7732558139534884, 'f1': 0.6352238805970151, 'number': 688}, 'overall_precision': 0.539906103286385, 'overall_recall': 0.6182795698924731, 'overall_f1': 0.5764411027568923, 'overall_accuracy': 0.8098098453883064}
			------------EPOCH 5---------------
Loss:  tensor(0.8694, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1706, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6374, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4557, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8475, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4888, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3575, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0441, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8450, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5164, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5740, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1686, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0222, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5741, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6794, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5947, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4546, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5187, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0881, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2710, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4754, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4655, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3437, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4850, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1592, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6535, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1933, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3683, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3702, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1714, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3107, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4613, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4575, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.49382716049382713, 'recall': 0.5607476635514018, 'f1': 0.5251641137855579, 'number': 428}, 'P': {'precision': 0.6317934782608695, 'recall': 0.6758720930232558, 'f1': 0.6530898876404493, 'number': 688}, 'overall_precision': 0.5769230769230769, 'overall_recall': 0.6317204301075269, 'overall_f1': 0.6030795551753636, 'overall_accuracy': 0.8306024524613471}
			------------EPOCH 6---------------
Loss:  tensor(0.5775, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7590, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3213, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4243, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0607, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1786, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3592, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5912, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3307, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8407, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6371, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3593, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3781, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0324, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4649, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5824, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5515, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6696, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3751, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0561, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1630, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2786, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3779, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3136, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1796, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6885, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3345, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3170, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2694, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1368, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2608, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2792, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3462, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4284, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.41735537190082644, 'recall': 0.7079439252336449, 'f1': 0.5251299826689776, 'number': 428}, 'P': {'precision': 0.6644182124789207, 'recall': 0.5726744186046512, 'f1': 0.615144418423107, 'number': 688}, 'overall_precision': 0.5284306292645944, 'overall_recall': 0.6245519713261649, 'overall_f1': 0.5724845995893224, 'overall_accuracy': 0.8020614892482673}
			------------EPOCH 7---------------
Loss:  tensor(0.7111, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7344, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3853, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2778, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0315, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2410, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5416, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2845, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3615, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7726, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7338, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4575, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1484, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3335, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4840, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3919, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5662, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2309, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3497, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0643, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1862, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3990, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3280, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2123, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2546, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5694, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1141, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2228, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2733, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0982, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2793, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2993, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3814, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4754, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5032258064516129, 'recall': 0.5467289719626168, 'f1': 0.5240761478163495, 'number': 428}, 'P': {'precision': 0.6186440677966102, 'recall': 0.7427325581395349, 'f1': 0.6750330250990754, 'number': 688}, 'overall_precision': 0.5770720371804803, 'overall_recall': 0.6675627240143369, 'overall_f1': 0.6190278354798504, 'overall_accuracy': 0.8200817487115692}
			------------EPOCH 8---------------
Loss:  tensor(0.3663, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6294, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2372, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3114, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2148, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0214, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1701, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5802, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2605, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1676, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5231, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2394, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2680, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0837, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1620, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4116, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3464, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5328, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1876, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1864, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0689, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1676, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1536, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1168, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1433, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4704, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0844, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1913, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1598, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1188, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1757, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1395, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1313, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3202, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.521551724137931, 'recall': 0.5654205607476636, 'f1': 0.5426008968609866, 'number': 428}, 'P': {'precision': 0.6401028277634961, 'recall': 0.7238372093023255, 'f1': 0.679399727148704, 'number': 688}, 'overall_precision': 0.5958132045088567, 'overall_recall': 0.6630824372759857, 'overall_f1': 0.6276505513146734, 'overall_accuracy': 0.8418695574906699}
			------------EPOCH 9---------------
Loss:  tensor(0.2264, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5269, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1407, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2210, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1736, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0222, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1471, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4694, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1353, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1701, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4734, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1481, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2487, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0823, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1933, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3746, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1578, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1157, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0371, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0513, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1102, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0710, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0939, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0821, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1447, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5368, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0470, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0870, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1163, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0775, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1085, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0607, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1478, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3174, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.49588477366255146, 'recall': 0.5630841121495327, 'f1': 0.5273522975929977, 'number': 428}, 'P': {'precision': 0.6035353535353535, 'recall': 0.6947674418604651, 'f1': 0.6459459459459459, 'number': 688}, 'overall_precision': 0.5625978090766823, 'overall_recall': 0.6442652329749103, 'overall_f1': 0.6006683375104428, 'overall_accuracy': 0.8354007464012796}
			------------EPOCH 10---------------
Loss:  tensor(0.1926, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4750, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0855, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1843, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1155, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0657, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0773, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3613, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0824, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0854, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4877, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3578, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0686, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1694, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0662, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0625, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3295, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2531, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0617, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0703, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0276, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0311, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0676, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0690, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0814, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0596, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1127, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0703, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0842, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0439, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0603, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0645, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0506, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2865, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5279642058165548, 'recall': 0.5514018691588785, 'f1': 0.5394285714285714, 'number': 428}, 'P': {'precision': 0.6645244215938303, 'recall': 0.751453488372093, 'f1': 0.7053206002728513, 'number': 688}, 'overall_precision': 0.6146938775510205, 'overall_recall': 0.6747311827956989, 'overall_f1': 0.6433148227253311, 'overall_accuracy': 0.8407677270303892}
			------------EPOCH 11---------------
Loss:  tensor(0.1252, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4868, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0508, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1231, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0666, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0488, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0441, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3331, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0814, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4463, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2969, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1492, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0566, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0414, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2740, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1975, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0447, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0157, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0536, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0328, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0492, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0333, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3657, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0580, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0982, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0645, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0484, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0551, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0446, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0428, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2028, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5369978858350951, 'recall': 0.5934579439252337, 'f1': 0.5638179800221975, 'number': 428}, 'P': {'precision': 0.6958333333333333, 'recall': 0.7281976744186046, 'f1': 0.7116477272727272, 'number': 688}, 'overall_precision': 0.6328583403185247, 'overall_recall': 0.6765232974910395, 'overall_f1': 0.6539627544391512, 'overall_accuracy': 0.8462413364137196}
			------------EPOCH 12---------------
Loss:  tensor(0.1133, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3902, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0337, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0907, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0435, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0403, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0283, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3463, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0198, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0643, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4590, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2507, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0740, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1571, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0486, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0635, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2711, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1713, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0310, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0377, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0120, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0381, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0328, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0434, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0227, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3413, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0352, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0292, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0342, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0571, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0400, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0160, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0274, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2089, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4827586206896552, 'recall': 0.6214953271028038, 'f1': 0.54341164453524, 'number': 428}, 'P': {'precision': 0.7085271317829457, 'recall': 0.6642441860465116, 'f1': 0.6856714178544636, 'number': 688}, 'overall_precision': 0.6045150501672241, 'overall_recall': 0.6478494623655914, 'overall_f1': 0.6254325259515572, 'overall_accuracy': 0.8311711391505242}
			------------EPOCH 13---------------
Loss:  tensor(0.1017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3846, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0265, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0363, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0564, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0276, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3273, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0404, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0602, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2316, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0434, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0845, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0467, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0265, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1727, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0193, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0308, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0174, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0455, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0359, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0424, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0197, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3573, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0154, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0231, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0482, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0423, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1783, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5379609544468547, 'recall': 0.5794392523364486, 'f1': 0.5579302587176602, 'number': 428}, 'P': {'precision': 0.6707988980716253, 'recall': 0.7078488372093024, 'f1': 0.6888260254596889, 'number': 688}, 'overall_precision': 0.6192080876158382, 'overall_recall': 0.6586021505376344, 'overall_f1': 0.6382978723404255, 'overall_accuracy': 0.8369646347965168}
			------------EPOCH 14---------------
Loss:  tensor(0.0626, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3670, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0154, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0868, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0252, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0305, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0153, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2911, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0181, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0707, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3841, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2166, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0294, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0760, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0436, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0199, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1855, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1670, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1467, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0175, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0176, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0331, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0230, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0370, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0280, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0247, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0704, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0230, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0225, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2535, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5458937198067633, 'recall': 0.5280373831775701, 'f1': 0.5368171021377671, 'number': 428}, 'P': {'precision': 0.6466992665036675, 'recall': 0.7688953488372093, 'f1': 0.7025232403718459, 'number': 688}, 'overall_precision': 0.6128246753246753, 'overall_recall': 0.6765232974910395, 'overall_f1': 0.6431005110732537, 'overall_accuracy': 0.8407677270303892}
			------------EPOCH 15---------------
Loss:  tensor(0.0453, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4670, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0800, 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(0.0299, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2485, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0228, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3734, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0199, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0669, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0522, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0160, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2382, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1273, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1864, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0150, 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(0.0049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0303, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0173, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0314, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0855, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3225, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0112, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0181, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0198, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0118, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0536, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0136, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0199, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2384, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5022026431718062, 'recall': 0.5327102803738317, 'f1': 0.5170068027210883, 'number': 428}, 'P': {'precision': 0.6619354838709678, 'recall': 0.7456395348837209, 'f1': 0.7012987012987013, 'number': 688}, 'overall_precision': 0.6029292107404394, 'overall_recall': 0.6639784946236559, 'overall_f1': 0.6319829424307035, 'overall_accuracy': 0.8443220188377466}
			------------EPOCH 16---------------
Loss:  tensor(0.0391, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3515, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0662, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0280, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2700, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0180, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0725, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3654, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2090, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0301, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0624, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0305, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1631, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1751, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0087, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0127, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0150, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0183, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3789, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0198, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0471, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0173, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1751, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.46705426356589147, 'recall': 0.5630841121495327, 'f1': 0.510593220338983, 'number': 428}, 'P': {'precision': 0.6587412587412588, 'recall': 0.684593023255814, 'f1': 0.6714183891660727, 'number': 688}, 'overall_precision': 0.578391551584078, 'overall_recall': 0.6379928315412187, 'overall_f1': 0.6067319982956967, 'overall_accuracy': 0.8314554824951128}
			------------EPOCH 17---------------
Loss:  tensor(0.0333, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3398, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0556, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0090, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0084, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2310, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0466, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3450, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1925, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0320, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0546, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0330, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0118, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1738, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1385, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0167, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0684, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2809, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0105, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0149, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0328, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0196, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0210, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1605, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.49691991786447637, 'recall': 0.5654205607476636, 'f1': 0.5289617486338798, 'number': 428}, 'P': {'precision': 0.6267880364109233, 'recall': 0.7005813953488372, 'f1': 0.661633493479753, 'number': 688}, 'overall_precision': 0.5764331210191083, 'overall_recall': 0.6487455197132617, 'overall_f1': 0.6104553119730186, 'overall_accuracy': 0.8368580060422961}
			------------EPOCH 18---------------
Loss:  tensor(0.0339, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3145, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0385, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0080, 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(0.0069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2144, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0274, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3184, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1688, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0208, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0382, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0148, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1224, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0210, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0160, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0198, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0184, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0176, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0804, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2731, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0085, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0155, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0090, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0335, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0185, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1412, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.45126353790613716, 'recall': 0.5841121495327103, 'f1': 0.5091649694501019, 'number': 428}, 'P': {'precision': 0.5703125, 'recall': 0.6366279069767442, 'f1': 0.6016483516483516, 'number': 688}, 'overall_precision': 0.5204236006051437, 'overall_recall': 0.6164874551971327, 'overall_f1': 0.5643970467596391, 'overall_accuracy': 0.8281855340323441}
			------------EPOCH 19---------------
Loss:  tensor(0.0447, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0202, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0735, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0184, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0084, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1852, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0441, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2693, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1548, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0157, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0247, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1130, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1241, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0040, 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(0.0139, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0634, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2480, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0150, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0175, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0313, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0150, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0111, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1324, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.45436507936507936, 'recall': 0.5350467289719626, 'f1': 0.49141630901287553, 'number': 428}, 'P': {'precision': 0.5624256837098692, 'recall': 0.6875, 'f1': 0.6187050359712231, 'number': 688}, 'overall_precision': 0.5219330855018587, 'overall_recall': 0.6290322580645161, 'overall_f1': 0.5704997968305566, 'overall_accuracy': 0.8415496712280078}
			------------EPOCH 20---------------
Loss:  tensor(0.0260, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2667, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0159, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0460, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1937, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0165, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0343, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2542, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1335, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0120, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0265, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0344, 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(0.1085, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0925, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0170, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0318, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0652, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2270, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0305, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1231, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.42857142857142855, 'recall': 0.5257009345794392, 'f1': 0.47219307450157394, 'number': 428}, 'P': {'precision': 0.5712545676004872, 'recall': 0.6816860465116279, 'f1': 0.6216037110669318, 'number': 688}, 'overall_precision': 0.5156017830609212, 'overall_recall': 0.6218637992831542, 'overall_f1': 0.5637692932575142, 'overall_accuracy': 0.8422605295894793}
			------------EPOCH 21---------------
Loss:  tensor(0.0175, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2466, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0128, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0467, 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(0.0210, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0201, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1784, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0098, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0441, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2499, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0204, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0185, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0877, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0943, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0761, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0111, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0306, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0259, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0686, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2538, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0176, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1079, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4543478260869565, 'recall': 0.4883177570093458, 'f1': 0.47072072072072074, 'number': 428}, 'P': {'precision': 0.649810366624526, 'recall': 0.747093023255814, 'f1': 0.6950642325895875, 'number': 688}, 'overall_precision': 0.5779376498800959, 'overall_recall': 0.6478494623655914, 'overall_f1': 0.6108998732572876, 'overall_accuracy': 0.8431135596232451}
			------------EPOCH 22---------------
Loss:  tensor(0.0161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2315, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0342, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0155, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1535, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0442, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2277, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1305, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0182, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0299, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0207, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1181, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0861, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0115, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0260, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0617, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0126, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0283, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0358, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0143, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1484, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5319634703196348, 'recall': 0.544392523364486, 'f1': 0.5381062355658198, 'number': 428}, 'P': {'precision': 0.6428571428571429, 'recall': 0.7063953488372093, 'f1': 0.6731301939058172, 'number': 688}, 'overall_precision': 0.602177554438861, 'overall_recall': 0.6442652329749103, 'overall_f1': 0.6225108225108226, 'overall_accuracy': 0.8362182335169718}
			------------EPOCH 23---------------
Loss:  tensor(0.0162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2459, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0284, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0196, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0077, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0569, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2680, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0228, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1129, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0879, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0152, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0241, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0525, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0668, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0463, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0296, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1477, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5472972972972973, 'recall': 0.5677570093457944, 'f1': 0.5573394495412844, 'number': 428}, 'P': {'precision': 0.6684141546526867, 'recall': 0.7412790697674418, 'f1': 0.7029634734665747, 'number': 688}, 'overall_precision': 0.623860811930406, 'overall_recall': 0.6747311827956989, 'overall_f1': 0.6482996125699526, 'overall_accuracy': 0.8473076239559267}
			------------EPOCH 24---------------
Loss:  tensor(0.0133, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2421, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0323, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1500, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0385, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2351, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0419, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0303, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1251, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0953, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0937, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, 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(0.0109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0121, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0471, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1865, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0233, 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(0.0128, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1223, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4906054279749478, 'recall': 0.5490654205607477, 'f1': 0.5181918412348402, 'number': 428}, 'P': {'precision': 0.6518324607329843, 'recall': 0.7238372093023255, 'f1': 0.6859504132231405, 'number': 688}, 'overall_precision': 0.589702333065165, 'overall_recall': 0.6568100358422939, 'overall_f1': 0.6214497668503604, 'overall_accuracy': 0.8323795983650257}
			------------EPOCH 25---------------
Loss:  tensor(0.0228, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2223, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0608, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1297, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0370, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1861, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0860, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0149, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0663, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0594, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0279, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0126, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0471, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0275, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0808, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5073375262054507, 'recall': 0.5654205607476636, 'f1': 0.5348066298342541, 'number': 428}, 'P': {'precision': 0.6505376344086021, 'recall': 0.7034883720930233, 'f1': 0.6759776536312849, 'number': 688}, 'overall_precision': 0.5945945945945946, 'overall_recall': 0.6505376344086021, 'overall_f1': 0.6213093709884467, 'overall_accuracy': 0.8363248622711925}
			------------EPOCH 26---------------
Loss:  tensor(0.0378, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1813, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0199, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0268, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0161, 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(0.0139, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1516, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0466, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0746, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0175, 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(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0754, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0892, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0574, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0345, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0142, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0511, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1670, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0140, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0130, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0279, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0882, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5474613686534217, 'recall': 0.5794392523364486, 'f1': 0.5629965947786606, 'number': 428}, 'P': {'precision': 0.6742323097463284, 'recall': 0.7340116279069767, 'f1': 0.7028531663187195, 'number': 688}, 'overall_precision': 0.6264559068219634, 'overall_recall': 0.6747311827956989, 'overall_f1': 0.6496980155306298, 'overall_accuracy': 0.8432557312955393}
			------------EPOCH 27---------------
Loss:  tensor(0.0206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1568, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0140, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0240, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0087, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0294, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1525, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0204, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0959, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1233, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0665, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0460, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1531, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, 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(0.0021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0210, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0998, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5498721227621484, 'recall': 0.5023364485981309, 'f1': 0.5250305250305249, 'number': 428}, 'P': {'precision': 0.6481927710843374, 'recall': 0.7819767441860465, 'f1': 0.7088274044795784, 'number': 688}, 'overall_precision': 0.6167076167076168, 'overall_recall': 0.6747311827956989, 'overall_f1': 0.644415917843389, 'overall_accuracy': 0.8446774480184823}
			------------EPOCH 28---------------
Loss:  tensor(0.0082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1972, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0202, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0107, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1115, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0342, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1488, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0927, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0126, 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(0.0950, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0822, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0629, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0023, 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(0.0049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0084, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0663, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2142, 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(0.0026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0248, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0024, 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(0.0853, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5311827956989247, 'recall': 0.5771028037383178, 'f1': 0.5531914893617021, 'number': 428}, 'P': {'precision': 0.6631299734748011, 'recall': 0.7267441860465116, 'f1': 0.6934812760055478, 'number': 688}, 'overall_precision': 0.6127973748974569, 'overall_recall': 0.6693548387096774, 'overall_f1': 0.6398286937901498, 'overall_accuracy': 0.8408032699484628}
			------------EPOCH 29---------------
Loss:  tensor(0.0061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1527, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0177, 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(0.0133, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0280, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1691, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0902, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0366, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0143, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0487, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0806, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0317, 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(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0090, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0396, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1555, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, 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(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0344, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1250, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5536480686695279, 'recall': 0.602803738317757, 'f1': 0.5771812080536913, 'number': 428}, 'P': {'precision': 0.6834239130434783, 'recall': 0.7311046511627907, 'f1': 0.7064606741573033, 'number': 688}, 'overall_precision': 0.6331114808652246, 'overall_recall': 0.681899641577061, 'overall_f1': 0.6566005176876618, 'overall_accuracy': 0.840945441620757}
			------------EPOCH 30---------------
Loss:  tensor(0.0052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2247, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0321, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0132, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0224, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1410, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0599, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0196, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0126, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0697, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0793, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0510, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, 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(0.0034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0065, 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(0.0588, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1641, 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(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0186, 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(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0717, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.51440329218107, 'recall': 0.5841121495327103, 'f1': 0.5470459518599563, 'number': 428}, 'P': {'precision': 0.6693657219973009, 'recall': 0.7209302325581395, 'f1': 0.6941917424772567, 'number': 688}, 'overall_precision': 0.6079869600651997, 'overall_recall': 0.6684587813620072, 'overall_f1': 0.636790439607341, 'overall_accuracy': 0.8395948107339612}


		-------------RUN 4-----------
			------------EPOCH 1---------------
Loss:  tensor(3.2185, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2838, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8371, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0452, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2514, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6758, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0973, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6887, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1538, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5890, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5788, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3551, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3157, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3564, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6343, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2231, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2218, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1923, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7422, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0639, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1423, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0527, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9690, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3816, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4115, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6990, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2921, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7416, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2852, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7408, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3231, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2666, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7580, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0142, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1352, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.22549019607843138, 'recall': 0.515274949083503, 'f1': 0.3137011779293243, 'number': 491}, 'P': {'precision': 0.35833333333333334, 'recall': 0.10462287104622871, 'f1': 0.16195856873822975, 'number': 822}, 'overall_precision': 0.2488986784140969, 'overall_recall': 0.25818735719725816, 'overall_f1': 0.2534579439252336, 'overall_accuracy': 0.5482000692281066}
			------------EPOCH 2---------------
Loss:  tensor(1.8796, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1276, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4882, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5905, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6833, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4538, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6387, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4314, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3181, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5774, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0150, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4601, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5921, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1882, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9968, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3116, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3599, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3503, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4900, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2787, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3520, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4197, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2786, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3794, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8910, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3087, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2772, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1886, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2548, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2760, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2638, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0147, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7520, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1793, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0864, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2797971259509721, 'recall': 0.6741344195519349, 'f1': 0.39545997610513745, 'number': 491}, 'P': {'precision': 0.4318181818181818, 'recall': 0.18491484184914841, 'f1': 0.25894378194207834, 'number': 822}, 'overall_precision': 0.31465798045602605, 'overall_recall': 0.36785986290936784, 'overall_f1': 0.33918539325842695, 'overall_accuracy': 0.628562362986039}
			------------EPOCH 3---------------
Loss:  tensor(1.4968, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0744, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2387, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4628, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3819, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4598, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9628, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9536, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0891, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0930, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1175, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8811, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0451, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9922, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3750, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9190, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9228, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0889, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8986, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0218, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6483, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1940, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2504, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1148, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1364, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8670, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6180, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8505, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9897, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1870, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6895, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8921, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9649, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0764, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.41935483870967744, 'recall': 0.6089613034623218, 'f1': 0.4966777408637874, 'number': 491}, 'P': {'precision': 0.631917631917632, 'recall': 0.597323600973236, 'f1': 0.6141338336460288, 'number': 822}, 'overall_precision': 0.5302013422818792, 'overall_recall': 0.6016755521706016, 'overall_f1': 0.5636817695326436, 'overall_accuracy': 0.8013153340256144}
			------------EPOCH 4---------------
Loss:  tensor(0.9260, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0316, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7779, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2806, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2456, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7752, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7270, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7505, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4660, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7642, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0678, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6955, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6314, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7418, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7249, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2893, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6806, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6819, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6306, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7591, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4620, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9338, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1759, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9492, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0946, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4509, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5913, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8781, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5573, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4487, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6669, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0485, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.462738301559792, 'recall': 0.5437881873727087, 'f1': 0.5, 'number': 491}, 'P': {'precision': 0.477859778597786, 'recall': 0.6301703163017032, 'f1': 0.5435466946484785, 'number': 822}, 'overall_precision': 0.47260686333534013, 'overall_recall': 0.5978674790555979, 'overall_f1': 0.5279085406859448, 'overall_accuracy': 0.8190261913003346}
			------------EPOCH 5---------------
Loss:  tensor(0.8349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0273, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5341, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6622, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1911, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1935, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2610, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5254, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4813, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3406, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6474, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4936, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0433, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4524, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4288, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5395, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4876, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1902, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4154, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4728, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3796, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5182, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6987, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1297, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7687, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0560, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3195, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4518, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7233, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0766, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2612, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3376, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4525, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0221, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5605700712589073, 'recall': 0.48065173116089616, 'f1': 0.5175438596491229, 'number': 491}, 'P': {'precision': 0.5769230769230769, 'recall': 0.7299270072992701, 'f1': 0.644468313641246, 'number': 822}, 'overall_precision': 0.5722108145106092, 'overall_recall': 0.6367098248286367, 'overall_f1': 0.6027397260273971, 'overall_accuracy': 0.8199492327218184}
			------------EPOCH 6---------------
Loss:  tensor(0.8040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0247, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3511, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4991, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1425, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1531, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1924, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3492, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3434, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2255, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5150, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3225, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0302, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3499, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4213, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3400, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1452, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2227, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3208, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3503, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2568, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4211, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2843, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5491, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0442, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2331, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3693, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5930, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0707, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2691, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3470, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0220, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.42424242424242425, 'recall': 0.5417515274949084, 'f1': 0.47584973166368516, 'number': 491}, 'P': {'precision': 0.6277712952158693, 'recall': 0.6545012165450121, 'f1': 0.6408576533650981, 'number': 822}, 'overall_precision': 0.5417789757412399, 'overall_recall': 0.6123381568926123, 'overall_f1': 0.5749016803718269, 'overall_accuracy': 0.7985462097611631}
			------------EPOCH 7---------------
Loss:  tensor(0.5984, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2463, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1508, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1695, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4447, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6200, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1587, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2451, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0141, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3298, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4403, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4955, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1968, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2954, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4400, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4322, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2585, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3442, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5837, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0713, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4983, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5555, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5118, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1169, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3712, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2505, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3000, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0201, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3560606060606061, 'recall': 0.670061099796334, 'f1': 0.465017667844523, 'number': 491}, 'P': {'precision': 0.644122383252818, 'recall': 0.48661800486618007, 'f1': 0.5544005544005544, 'number': 822}, 'overall_precision': 0.47184466019417476, 'overall_recall': 0.5552170601675552, 'overall_f1': 0.510146955913226, 'overall_accuracy': 0.7404811353409484}
			------------EPOCH 8---------------
Loss:  tensor(0.9152, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0605, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5351, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6261, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2633, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1522, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3542, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2638, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3246, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1288, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4585, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2507, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0193, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2660, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3890, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2184, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2851, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3186, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1367, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2684, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2532, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3085, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1880, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2779, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4484, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0832, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1594, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1367, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2183, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4407, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0415, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1760, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1536, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0131, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.44598765432098764, 'recall': 0.5885947046843177, 'f1': 0.5074626865671642, 'number': 491}, 'P': {'precision': 0.6074534161490683, 'recall': 0.5948905109489051, 'f1': 0.6011063306699447, 'number': 822}, 'overall_precision': 0.5354439091534756, 'overall_recall': 0.5925361766945926, 'overall_f1': 0.5625451916124368, 'overall_accuracy': 0.8022672204915196}
			------------EPOCH 9---------------
Loss:  tensor(0.6190, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0112, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1763, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1835, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1142, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0925, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1684, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1838, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0470, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3552, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1176, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1742, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0857, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0717, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0996, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0956, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0869, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1087, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1595, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0893, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1938, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0511, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0786, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4491, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0819, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1588, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3459, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0118, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0709, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0745, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0046, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4634146341463415, 'recall': 0.5804480651731161, 'f1': 0.5153707052441231, 'number': 491}, 'P': {'precision': 0.6593406593406593, 'recall': 0.656934306569343, 'f1': 0.6581352833638026, 'number': 822}, 'overall_precision': 0.5753138075313807, 'overall_recall': 0.6283320639756284, 'overall_f1': 0.6006552602839461, 'overall_accuracy': 0.8135167878158532}
			------------EPOCH 10---------------
Loss:  tensor(0.3880, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0705, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0800, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0561, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1258, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0414, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0573, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1292, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3127, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0582, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0857, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2909, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0481, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0606, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0555, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0457, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0569, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0675, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0350, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0584, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3980, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0463, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0406, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1328, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0851, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0353, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0666, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0465, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0035, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4570446735395189, 'recall': 0.5417515274949084, 'f1': 0.49580615097856473, 'number': 491}, 'P': {'precision': 0.6372093023255814, 'recall': 0.6666666666666666, 'f1': 0.6516052318668253, 'number': 822}, 'overall_precision': 0.5644937586685159, 'overall_recall': 0.61995430312262, 'overall_f1': 0.5909255898366607, 'overall_accuracy': 0.8145552094150225}
			------------EPOCH 11---------------
Loss:  tensor(0.3368, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0507, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0627, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0456, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0330, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0326, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0290, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0992, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0143, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2762, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0344, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0370, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2560, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0244, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0198, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0237, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0180, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0233, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0303, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0610, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0350, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1422, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0252, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2493, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0578, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3830, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0320, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0997, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2504, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0397, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0328, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0421, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0284, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0029, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5019685039370079, 'recall': 0.5193482688391039, 'f1': 0.5105105105105106, 'number': 491}, 'P': {'precision': 0.5950323974082073, 'recall': 0.670316301703163, 'f1': 0.6304347826086956, 'number': 822}, 'overall_precision': 0.5620641562064156, 'overall_recall': 0.6138613861386139, 'overall_f1': 0.5868219876228613, 'overall_accuracy': 0.8205549786546671}
			------------EPOCH 12---------------
Loss:  tensor(0.3612, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0516, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0430, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0624, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0255, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0330, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0236, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1188, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0077, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2878, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0342, 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(0.2069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0143, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0239, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0392, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0114, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1382, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0191, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2352, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0707, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4212, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0297, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2600, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0436, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0265, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0507, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0023, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5512820512820513, 'recall': 0.5254582484725051, 'f1': 0.5380604796663192, 'number': 491}, 'P': {'precision': 0.6029411764705882, 'recall': 0.6982968369829684, 'f1': 0.6471251409244645, 'number': 822}, 'overall_precision': 0.5859154929577465, 'overall_recall': 0.6336633663366337, 'overall_f1': 0.6088547383827296, 'overall_accuracy': 0.8204107534325603}
			------------EPOCH 13---------------
Loss:  tensor(0.3354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0481, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0398, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1231, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0169, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0280, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0153, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0941, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2505, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0282, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0177, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1980, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0130, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0085, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0259, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0300, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0887, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1633, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0517, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3524, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0201, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0202, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0725, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2139, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0386, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0196, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0020, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4864376130198915, 'recall': 0.5478615071283096, 'f1': 0.5153256704980844, 'number': 491}, 'P': {'precision': 0.5508130081300813, 'recall': 0.6593673965936739, 'f1': 0.6002214839424141, 'number': 822}, 'overall_precision': 0.52765126870527, 'overall_recall': 0.6176694592536177, 'overall_f1': 0.5691228070175438, 'overall_accuracy': 0.8207568939656167}
			------------EPOCH 14---------------
Loss:  tensor(0.2534, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0395, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0391, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0331, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0977, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0098, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0211, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0076, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0914, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1862, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0165, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1703, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0153, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0212, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0393, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0672, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0102, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1397, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0466, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0112, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1588, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0370, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0244, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0126, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.36161879895561355, 'recall': 0.5641547861507128, 'f1': 0.4407319013524264, 'number': 491}, 'P': {'precision': 0.5573212258796821, 'recall': 0.597323600973236, 'f1': 0.5766294773928362, 'number': 822}, 'overall_precision': 0.4663023679417122, 'overall_recall': 0.584920030464585, 'overall_f1': 0.518918918918919, 'overall_accuracy': 0.8056997807776624}
			------------EPOCH 15---------------
Loss:  tensor(0.3671, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0448, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0383, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0322, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1225, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0463, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1369, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0164, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1720, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0074, 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(0.0045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0139, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0311, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1076, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1656, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0410, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3474, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0532, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1713, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0465, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.42319277108433734, 'recall': 0.5723014256619144, 'f1': 0.48658008658008656, 'number': 491}, 'P': {'precision': 0.5914110429447853, 'recall': 0.5863746958637469, 'f1': 0.5888821014050092, 'number': 822}, 'overall_precision': 0.5158891142663962, 'overall_recall': 0.5811119573495811, 'overall_f1': 0.5465616045845272, 'overall_accuracy': 0.7988058151609554}
			------------EPOCH 16---------------
Loss:  tensor(0.3777, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0243, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0251, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0427, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1164, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0160, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0699, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2485, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0139, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1960, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0112, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0174, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0976, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1616, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0090, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0615, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2186, 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(0.0042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0251, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.44112769485903813, 'recall': 0.5417515274949084, 'f1': 0.48628884826325414, 'number': 491}, 'P': {'precision': 0.6020293122886133, 'recall': 0.6496350364963503, 'f1': 0.6249268578115857, 'number': 822}, 'overall_precision': 0.5369127516778524, 'overall_recall': 0.6092916984006093, 'overall_f1': 0.5708169818052087, 'overall_accuracy': 0.8130841121495327}
			------------EPOCH 17---------------
Loss:  tensor(0.1973, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0174, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0707, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0139, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0840, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2144, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0205, 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>)
Loss:  tensor(0.1985, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0077, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0037, 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(0.0154, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0195, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0102, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0572, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1142, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0360, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2674, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0182, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0506, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1615, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0439, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0336, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5089820359281437, 'recall': 0.5193482688391039, 'f1': 0.5141129032258065, 'number': 491}, 'P': {'precision': 0.5617103984450923, 'recall': 0.7031630170316302, 'f1': 0.6245272825499729, 'number': 822}, 'overall_precision': 0.5444444444444444, 'overall_recall': 0.6344249809596344, 'overall_f1': 0.5860007034822371, 'overall_accuracy': 0.821795315564786}
			------------EPOCH 18---------------
Loss:  tensor(0.2087, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0087, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0171, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0181, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0706, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0123, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0590, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1465, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0191, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0137, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1479, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0197, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0128, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0136, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0517, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0321, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2590, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0556, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0369, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0165, 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>)
				 {'C': {'precision': 0.47619047619047616, 'recall': 0.5295315682281059, 'f1': 0.5014464802314368, 'number': 491}, 'P': {'precision': 0.5631951466127402, 'recall': 0.6776155717761557, 'f1': 0.6151297625621204, 'number': 822}, 'overall_precision': 0.5322475570032573, 'overall_recall': 0.6222391469916222, 'overall_f1': 0.5737359550561797, 'overall_accuracy': 0.8215068651205722}
			------------EPOCH 19---------------
Loss:  tensor(0.1923, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0238, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0650, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0454, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1439, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0154, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1338, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0398, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0287, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0098, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0123, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0420, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0759, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0335, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2457, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0283, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0986, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0335, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0271, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0157, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0114, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4677137870855148, 'recall': 0.5458248472505092, 'f1': 0.5037593984962405, 'number': 491}, 'P': {'precision': 0.6367869615832363, 'recall': 0.6654501216545012, 'f1': 0.6508030933967877, 'number': 822}, 'overall_precision': 0.5691340782122905, 'overall_recall': 0.6207159177456207, 'overall_f1': 0.5938069216757741, 'overall_accuracy': 0.8143532941040729}
			------------EPOCH 20---------------
Loss:  tensor(0.2212, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0136, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0595, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0174, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0328, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0839, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0087, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0115, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0537, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0899, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0337, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0338, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0433, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0145, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0115, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4685430463576159, 'recall': 0.5763747454175153, 'f1': 0.5168949771689499, 'number': 491}, 'P': {'precision': 0.6515337423312884, 'recall': 0.6459854014598541, 'f1': 0.6487477092241906, 'number': 822}, 'overall_precision': 0.5736434108527132, 'overall_recall': 0.61995430312262, 'overall_f1': 0.5959004392386531, 'overall_accuracy': 0.8104015230183454}
			------------EPOCH 21---------------
Loss:  tensor(0.2140, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0067, 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(0.0978, 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(0.0076, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0476, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1510, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0169, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1325, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0027, 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(0.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0090, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0301, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0701, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0225, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0933, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0331, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.47769516728624534, 'recall': 0.5234215885947047, 'f1': 0.4995140913508261, 'number': 491}, 'P': {'precision': 0.5782038345105953, 'recall': 0.6970802919708029, 'f1': 0.6321014892443463, 'number': 822}, 'overall_precision': 0.5428384565075213, 'overall_recall': 0.6321401370906321, 'overall_f1': 0.5840957072484166, 'overall_accuracy': 0.8196607822776047}
			------------EPOCH 22---------------
Loss:  tensor(0.1473, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0614, 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(0.0035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0137, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0700, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0118, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0183, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0875, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0015, 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(0.0512, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0237, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1656, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0608, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0323, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0094, 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>)
				 {'C': {'precision': 0.4906015037593985, 'recall': 0.5315682281059063, 'f1': 0.5102639296187683, 'number': 491}, 'P': {'precision': 0.5933952528379773, 'recall': 0.6995133819951338, 'f1': 0.6420993858179788, 'number': 822}, 'overall_precision': 0.5569620253164557, 'overall_recall': 0.6367098248286367, 'overall_f1': 0.5941719971570718, 'overall_accuracy': 0.8209876543209876}
			------------EPOCH 23---------------
Loss:  tensor(0.1123, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.4200e-05, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0447, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0672, 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(0.0083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0745, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0048, 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(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0318, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0181, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1407, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, 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(0.0072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0530, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0068, 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>)
				 {'C': {'precision': 0.4805914972273567, 'recall': 0.5295315682281059, 'f1': 0.5038759689922481, 'number': 491}, 'P': {'precision': 0.5989528795811518, 'recall': 0.6958637469586375, 'f1': 0.6437816544738323, 'number': 822}, 'overall_precision': 0.5561497326203209, 'overall_recall': 0.6336633663366337, 'overall_f1': 0.592381630473478, 'overall_accuracy': 0.8189685012114919}
			------------EPOCH 24---------------
Loss:  tensor(0.0970, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.5015e-05, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0331, 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(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0098, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0555, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0034, 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(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0267, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0149, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1216, 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(0.0068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0445, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0327, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0061, 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>)
				 {'C': {'precision': 0.44867549668874174, 'recall': 0.5519348268839104, 'f1': 0.4949771689497717, 'number': 491}, 'P': {'precision': 0.6030871003307607, 'recall': 0.6654501216545012, 'f1': 0.6327356853672643, 'number': 822}, 'overall_precision': 0.5413633355393779, 'overall_recall': 0.623000761614623, 'overall_f1': 0.5793201133144477, 'overall_accuracy': 0.8153340256143995}
			------------EPOCH 25---------------
Loss:  tensor(0.0890, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.7417e-05, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0402, 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(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0330, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0076, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0493, 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(0.0011, 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(0.0060, 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(0.0131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0286, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1337, 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(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0395, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0340, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0001, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.44922547332185886, 'recall': 0.5315682281059063, 'f1': 0.4869402985074626, 'number': 491}, 'P': {'precision': 0.5930735930735931, 'recall': 0.6666666666666666, 'f1': 0.6277205040091638, 'number': 822}, 'overall_precision': 0.5375415282392026, 'overall_recall': 0.6161462300076161, 'overall_f1': 0.5741660752306601, 'overall_accuracy': 0.816632052613361}
			------------EPOCH 26---------------
Loss:  tensor(0.0711, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.1333e-05, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0251, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, 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(0.0073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0277, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0619, 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(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0123, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0201, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1462, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0001, 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(0.0037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0307, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0319, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0001, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5337690631808278, 'recall': 0.4989816700610998, 'f1': 0.5157894736842105, 'number': 491}, 'P': {'precision': 0.5838206627680312, 'recall': 0.7287104622871047, 'f1': 0.6482683982683983, 'number': 822}, 'overall_precision': 0.5683501683501684, 'overall_recall': 0.6428027418126429, 'overall_f1': 0.603288062902073, 'overall_accuracy': 0.8216799353871005}
			------------EPOCH 27---------------
Loss:  tensor(0.1411, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.7737e-05, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, 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(0.0082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0367, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0533, 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(0.0010, 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(0.0060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, 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(0.0261, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1354, 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(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0084, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0529, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0043, 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>)
				 {'C': {'precision': 0.46702317290552586, 'recall': 0.5336048879837068, 'f1': 0.4980988593155894, 'number': 491}, 'P': {'precision': 0.6010752688172043, 'recall': 0.6800486618004866, 'f1': 0.6381278538812786, 'number': 822}, 'overall_precision': 0.5506371562709591, 'overall_recall': 0.6252856054836253, 'overall_f1': 0.5855920114122681, 'overall_accuracy': 0.8170935733241029}
			------------EPOCH 28---------------
Loss:  tensor(0.0774, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.1119e-05, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, 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(0.0196, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0106, 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(0.0075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0482, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, 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(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0268, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0210, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1154, 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(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0085, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0474, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0204, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0159, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0001, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.42700156985871274, 'recall': 0.5539714867617108, 'f1': 0.4822695035460993, 'number': 491}, 'P': {'precision': 0.5855555555555556, 'recall': 0.6411192214111923, 'f1': 0.6120789779326365, 'number': 822}, 'overall_precision': 0.5198438516590761, 'overall_recall': 0.6085300837776085, 'overall_f1': 0.5607017543859649, 'overall_accuracy': 0.8120456905503635}
			------------EPOCH 29---------------
Loss:  tensor(0.0891, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4415e-05, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0084, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0434, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0624, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0093, 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(0.0400, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, 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(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0105, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0277, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0154, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0121, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0364, 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(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0001, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.46621621621621623, 'recall': 0.5621181262729125, 'f1': 0.5096952908587258, 'number': 491}, 'P': {'precision': 0.5906623235613464, 'recall': 0.6618004866180048, 'f1': 0.6242111302352267, 'number': 822}, 'overall_precision': 0.541969596827495, 'overall_recall': 0.6245239908606245, 'overall_f1': 0.5803255484784147, 'overall_accuracy': 0.8198626975885542}
			------------EPOCH 30---------------
Loss:  tensor(0.0597, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3313e-05, 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(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0186, 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(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0587, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.9074e-05, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0372, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, 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(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, 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(0.0083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0202, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0085, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0701, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.3597e-05, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0271, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, 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(8.7560e-05, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.483695652173913, 'recall': 0.5437881873727087, 'f1': 0.5119846596356663, 'number': 491}, 'P': {'precision': 0.5959703075291622, 'recall': 0.683698296836983, 'f1': 0.6368271954674222, 'number': 822}, 'overall_precision': 0.5545150501672241, 'overall_recall': 0.6313785224676314, 'overall_f1': 0.5904558404558404, 'overall_accuracy': 0.8197184723664475}


		-------------RUN 5-----------
			------------EPOCH 1---------------
Loss:  tensor(3.5581, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9779, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7827, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9766, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8819, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5729, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0783, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7871, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4873, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6376, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5188, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1936, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5102, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7462, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1864, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4204, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3295, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0510, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2876, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6425, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8187, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5286, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8538, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0322, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1391, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9645, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0242, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2908, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3459, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6626, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3796, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1144, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2223, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6298, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2832, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.1752136752136752, 'recall': 0.07008547008547009, 'f1': 0.10012210012210011, 'number': 585}, 'P': {'precision': 0.4668608885651857, 'recall': 0.6520854526958291, 'f1': 0.5441426146010186, 'number': 983}, 'overall_precision': 0.42439327940261357, 'overall_recall': 0.43494897959183676, 'overall_f1': 0.4296062992125984, 'overall_accuracy': 0.7272961351288719}
			------------EPOCH 2---------------
Loss:  tensor(1.6591, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7311, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5607, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5244, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5815, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3403, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3976, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0709, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6175, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4659, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5168, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0858, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4361, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9807, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3470, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4496, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1808, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0252, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3808, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1885, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3498, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5666, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4786, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2356, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1997, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1894, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2185, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2994, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1261, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7792, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8894, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2311, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2763, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3236842105263158, 'recall': 0.21025641025641026, 'f1': 0.2549222797927461, 'number': 585}, 'P': {'precision': 0.4370515329419439, 'recall': 0.681586978636826, 'f1': 0.5325914149443561, 'number': 983}, 'overall_precision': 0.4145321484579195, 'overall_recall': 0.5057397959183674, 'overall_f1': 0.4556162022407354, 'overall_accuracy': 0.758966964492623}
			------------EPOCH 3---------------
Loss:  tensor(1.1652, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4802, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2210, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2609, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9196, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0320, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7437, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1866, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2211, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0863, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1869, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0518, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0518, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7288, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0865, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9276, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1901, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7574, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1210, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9633, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2956, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4374, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2667, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9928, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9087, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1689, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9223, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1475, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1741, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9635, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8239, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5696, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7202, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0293, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2325, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3508771929824561, 'recall': 0.5811965811965812, 'f1': 0.4375804375804376, 'number': 585}, 'P': {'precision': 0.544904137235116, 'recall': 0.5493387589013224, 'f1': 0.547112462006079, 'number': 983}, 'overall_precision': 0.4489795918367347, 'overall_recall': 0.5612244897959183, 'overall_f1': 0.4988662131519274, 'overall_accuracy': 0.7991090969951335}
			------------EPOCH 4---------------
Loss:  tensor(0.9205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3612, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9839, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0207, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6795, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8177, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6377, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9358, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7610, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0553, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8545, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0594, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7493, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8487, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8874, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0685, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5686, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6290, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8957, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7873, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2511, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3532, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0900, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7951, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7315, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9722, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6636, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1169, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8574, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6891, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1897, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3975, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8501, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2159, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4830508474576271, 'recall': 0.48717948717948717, 'f1': 0.4851063829787234, 'number': 585}, 'P': {'precision': 0.5958677685950413, 'recall': 0.7334689725330621, 'f1': 0.657546739626083, 'number': 983}, 'overall_precision': 0.5588888888888889, 'overall_recall': 0.6415816326530612, 'overall_f1': 0.5973871733966746, 'overall_accuracy': 0.8292607564950949}
			------------EPOCH 5---------------
Loss:  tensor(0.6738, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2693, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9440, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2430, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5157, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4637, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5964, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7770, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5629, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6693, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0308, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6184, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4420, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0284, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4988, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5548, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7840, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0424, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3980, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1852, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2337, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9437, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5523, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5160, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7372, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4400, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0790, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0860, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6568, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1518, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2565, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5477, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6812, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1916, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4659248956884562, 'recall': 0.5726495726495726, 'f1': 0.513803680981595, 'number': 585}, 'P': {'precision': 0.6472222222222223, 'recall': 0.711088504577823, 'f1': 0.6776539020843432, 'number': 983}, 'overall_precision': 0.574763757643135, 'overall_recall': 0.6594387755102041, 'overall_f1': 0.6141966141966142, 'overall_accuracy': 0.8317583747457321}
			------------EPOCH 6---------------
Loss:  tensor(0.4297, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2313, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6891, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8293, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1801, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3529, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4337, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6793, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3634, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0300, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2924, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0168, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3460, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2927, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2634, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3967, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2941, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1454, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7505, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3159, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3598, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5827, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2824, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0756, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0673, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4649, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3173, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1236, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1534, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4177, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1790, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4214360041623309, 'recall': 0.6923076923076923, 'f1': 0.5239327296248383, 'number': 585}, 'P': {'precision': 0.6643835616438356, 'recall': 0.5920651068158698, 'f1': 0.6261430876815492, 'number': 983}, 'overall_precision': 0.5372890582471421, 'overall_recall': 0.6294642857142857, 'overall_f1': 0.5797356828193833, 'overall_accuracy': 0.8124468934263718}
			------------EPOCH 7---------------
Loss:  tensor(0.3722, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2311, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6287, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6960, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1374, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2362, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3579, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2879, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2535, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4239, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0251, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2935, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2906, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0173, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5550, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6909, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0453, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2197, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2767, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3963, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3551, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1490, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1269, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7851, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2953, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3835, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6377, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4875, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0696, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4312, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2934, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1231, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1668, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3567, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5361, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1072, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.398406374501992, 'recall': 0.5128205128205128, 'f1': 0.44843049327354256, 'number': 585}, 'P': {'precision': 0.6659459459459459, 'recall': 0.6266531027466938, 'f1': 0.6457023060796646, 'number': 983}, 'overall_precision': 0.5458879618593564, 'overall_recall': 0.5841836734693877, 'overall_f1': 0.5643869377695625, 'overall_accuracy': 0.8003192831578134}
			------------EPOCH 8---------------
Loss:  tensor(0.7429, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9355, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1789, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4420, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4809, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2573, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5523, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3328, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5475, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0521, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4893, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5745, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8682, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0378, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4510, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2643, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1904, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4806, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5502, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8085, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4111, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0456, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0840, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4734, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2941, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1148, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2374, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4741, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6300, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2097, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5307692307692308, 'recall': 0.4717948717948718, 'f1': 0.4995475113122172, 'number': 585}, 'P': {'precision': 0.6390076988879384, 'recall': 0.7599186164801628, 'f1': 0.6942379182156133, 'number': 983}, 'overall_precision': 0.605683836589698, 'overall_recall': 0.6524234693877551, 'overall_f1': 0.6281854467301198, 'overall_accuracy': 0.8286427890928754}
			------------EPOCH 9---------------
Loss:  tensor(0.3709, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1819, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5739, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2615, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2493, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1671, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2577, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4754, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2600, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3886, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0163, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2090, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1411, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0116, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1395, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1628, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4776, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0142, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1925, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1629, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2380, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1482, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1237, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0813, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6804, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1254, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1404, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4609, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0223, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2355, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0979, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0758, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0643, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2953, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3494, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0956, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5009174311926605, 'recall': 0.4666666666666667, 'f1': 0.4831858407079646, 'number': 585}, 'P': {'precision': 0.6143459915611814, 'recall': 0.7405900305188199, 'f1': 0.6715867158671587, 'number': 983}, 'overall_precision': 0.5786127167630057, 'overall_recall': 0.6383928571428571, 'overall_f1': 0.6070345664038811, 'overall_accuracy': 0.8312948991940675}
			------------EPOCH 10---------------
Loss:  tensor(0.1656, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1195, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4643, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5227, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0497, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1702, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1620, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0716, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3560, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0863, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2364, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0958, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0938, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4937, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1518, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1128, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2765, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1668, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1293, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0495, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5802, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0845, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0908, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0754, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1918, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0729, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0661, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0381, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2528, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0766, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5030181086519114, 'recall': 0.42735042735042733, 'f1': 0.4621072088724584, 'number': 585}, 'P': {'precision': 0.5930047694753577, 'recall': 0.7589013224821973, 'f1': 0.6657742079428827, 'number': 983}, 'overall_precision': 0.5675213675213675, 'overall_recall': 0.6352040816326531, 'overall_f1': 0.5994583207944628, 'overall_accuracy': 0.8283080567500064}
			------------EPOCH 11---------------
Loss:  tensor(0.0771, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0830, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3998, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4992, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0488, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0965, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0444, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1224, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3328, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0706, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0789, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0734, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0724, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0487, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3887, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0911, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0901, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1522, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0658, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1105, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0537, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5276, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0648, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0939, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0698, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1704, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0642, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0777, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0346, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2483, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3177, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0745, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.43676222596964587, 'recall': 0.4427350427350427, 'f1': 0.4397283531409168, 'number': 585}, 'P': {'precision': 0.5429657794676807, 'recall': 0.7263479145473042, 'f1': 0.6214099216710183, 'number': 983}, 'overall_precision': 0.509958071278826, 'overall_recall': 0.6205357142857143, 'overall_f1': 0.5598388952819332, 'overall_accuracy': 0.8255272034400185}
			------------EPOCH 12---------------
Loss:  tensor(0.0819, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0642, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3712, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4518, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0299, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0488, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0575, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0356, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0884, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3417, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0702, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2594, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1248, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0425, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0572, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0545, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4899, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0655, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0749, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0961, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0451, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1463, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4984, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0476, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0534, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2686, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0491, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0126, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1742, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0663, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0688, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0486, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2632, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3315, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0565, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.491679273827534, 'recall': 0.5555555555555556, 'f1': 0.521669341894061, 'number': 585}, 'P': {'precision': 0.5327476038338658, 'recall': 0.6785350966429298, 'f1': 0.5968680089485459, 'number': 983}, 'overall_precision': 0.5185572399372713, 'overall_recall': 0.6326530612244898, 'overall_f1': 0.5699511634587763, 'overall_accuracy': 0.8327883204160981}
			------------EPOCH 13---------------
Loss:  tensor(0.0663, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0710, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3836, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4767, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0169, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0562, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0500, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0368, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0407, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2728, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0332, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1668, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0357, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0334, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0478, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3473, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0627, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0727, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0861, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0368, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0141, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5672, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0473, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0511, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2734, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0412, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1143, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0271, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0651, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0123, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2220, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2468, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0467, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.496875, 'recall': 0.5435897435897435, 'f1': 0.5191836734693877, 'number': 585}, 'P': {'precision': 0.5730050933786078, 'recall': 0.6866734486266531, 'f1': 0.6247107820453494, 'number': 983}, 'overall_precision': 0.5462046204620462, 'overall_recall': 0.6332908163265306, 'overall_f1': 0.5865327820437094, 'overall_accuracy': 0.8332260473260036}
			------------EPOCH 14---------------
Loss:  tensor(0.0525, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0504, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3137, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0492, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0383, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0363, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0430, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3371, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0392, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1917, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0327, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0341, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0363, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0164, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3578, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0537, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0546, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0947, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4499, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0266, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0305, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2234, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0316, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0318, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0544, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2454, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2646, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0514, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.46143617021276595, 'recall': 0.5931623931623932, 'f1': 0.5190725504861631, 'number': 585}, 'P': {'precision': 0.6571428571428571, 'recall': 0.7019328585961343, 'f1': 0.6787998032464339, 'number': 983}, 'overall_precision': 0.5754716981132075, 'overall_recall': 0.6613520408163265, 'overall_f1': 0.6154302670623145, 'overall_accuracy': 0.8328913149831346}
			------------EPOCH 15---------------
Loss:  tensor(0.0371, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0414, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3000, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3897, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0197, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0231, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0148, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0260, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2223, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0333, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0275, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0418, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0727, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0275, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0642, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0454, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0241, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0137, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1118, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4303, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0187, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1817, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0177, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0773, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0322, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0444, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0213, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0333, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.44831880448318806, 'recall': 0.6153846153846154, 'f1': 0.5187319884726226, 'number': 585}, 'P': {'precision': 0.6832814122533749, 'recall': 0.669379450661241, 'f1': 0.6762589928057554, 'number': 983}, 'overall_precision': 0.5764439411098528, 'overall_recall': 0.6492346938775511, 'overall_f1': 0.6106778644271146, 'overall_accuracy': 0.8288745268687077}
			------------EPOCH 16---------------
Loss:  tensor(0.0758, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0397, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2802, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3657, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0105, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0198, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0129, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1958, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0183, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0926, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0170, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0276, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0156, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3191, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0338, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0398, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0274, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0838, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4428, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1449, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0147, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0725, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0450, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2084, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2496, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0462, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5211038961038961, 'recall': 0.5487179487179488, 'f1': 0.5345545378850958, 'number': 585}, 'P': {'precision': 0.6541705716963448, 'recall': 0.7100712105798576, 'f1': 0.6809756097560975, 'number': 983}, 'overall_precision': 0.6054664289958408, 'overall_recall': 0.6498724489795918, 'overall_f1': 0.6268840356813289, 'overall_accuracy': 0.8367793598887658}
			------------EPOCH 17---------------
Loss:  tensor(0.0616, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2468, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3371, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0077, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0202, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0132, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0175, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2306, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0186, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1197, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0266, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0165, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3700, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0419, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0618, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0308, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0201, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0918, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3976, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0167, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0127, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1255, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0182, 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(0.0626, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0324, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1586, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1863, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0350, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5764925373134329, 'recall': 0.5282051282051282, 'f1': 0.5512934879571811, 'number': 585}, 'P': {'precision': 0.6733393994540491, 'recall': 0.7527975584944049, 'f1': 0.7108549471661864, 'number': 983}, 'overall_precision': 0.6415902140672783, 'overall_recall': 0.6690051020408163, 'overall_f1': 0.6550109272556979, 'overall_accuracy': 0.8388907485130159}
			------------EPOCH 18---------------
Loss:  tensor(0.0326, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0271, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2558, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0166, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0148, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0160, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1535, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0193, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0748, 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(0.0155, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0321, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2562, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0148, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0380, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0152, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0693, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3570, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0171, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0200, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1105, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0177, 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(0.0629, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0390, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1516, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1814, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0351, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.49698795180722893, 'recall': 0.5641025641025641, 'f1': 0.5284227381905525, 'number': 585}, 'P': {'precision': 0.6533084808946877, 'recall': 0.7131230925737538, 'f1': 0.6819066147859922, 'number': 983}, 'overall_precision': 0.5935521013241221, 'overall_recall': 0.6575255102040817, 'overall_f1': 0.6239031770045387, 'overall_accuracy': 0.8335865283106316}
			------------EPOCH 19---------------
Loss:  tensor(0.0180, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0145, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1768, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2574, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0197, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1237, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0799, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0147, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0284, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0114, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2958, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0298, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0454, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0773, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0960, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0302, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0207, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1285, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0239, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4684813753581662, 'recall': 0.558974358974359, 'f1': 0.5097427903351519, 'number': 585}, 'P': {'precision': 0.6710650329877474, 'recall': 0.7243133265513734, 'f1': 0.6966731898238748, 'number': 983}, 'overall_precision': 0.5906765207504264, 'overall_recall': 0.6626275510204082, 'overall_f1': 0.6245867147580403, 'overall_accuracy': 0.8337152715194274}
			------------EPOCH 20---------------
Loss:  tensor(0.0257, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0114, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1806, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2640, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0956, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2214, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0159, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0257, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0531, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2655, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0626, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0186, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0175, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0962, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0202, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.49922239502332816, 'recall': 0.5487179487179488, 'f1': 0.5228013029315961, 'number': 585}, 'P': {'precision': 0.6747437092264679, 'recall': 0.7365208545269583, 'f1': 0.7042801556420233, 'number': 983}, 'overall_precision': 0.6089743589743589, 'overall_recall': 0.6664540816326531, 'overall_f1': 0.6364190012180267, 'overall_accuracy': 0.8362386384118238}
			------------EPOCH 21---------------
Loss:  tensor(0.0196, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1380, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0121, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0752, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0090, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0391, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0064, 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(0.0025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2085, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0406, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0498, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2463, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0533, 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(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0135, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0149, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0787, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0961, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0184, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5164473684210527, 'recall': 0.5367521367521367, 'f1': 0.5264040234702431, 'number': 585}, 'P': {'precision': 0.6660649819494585, 'recall': 0.7507629704984741, 'f1': 0.7058823529411765, 'number': 983}, 'overall_precision': 0.6130536130536131, 'overall_recall': 0.6709183673469388, 'overall_f1': 0.6406820950060902, 'overall_accuracy': 0.8373973272909854}
			------------EPOCH 22---------------
Loss:  tensor(0.0193, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1427, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0646, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0084, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0332, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0043, 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(0.0193, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1936, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0426, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0482, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2297, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0457, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0184, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0171, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0876, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0972, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0242, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4984567901234568, 'recall': 0.5521367521367522, 'f1': 0.5239253852392539, 'number': 585}, 'P': {'precision': 0.6626506024096386, 'recall': 0.7273652085452695, 'f1': 0.6935014548981571, 'number': 983}, 'overall_precision': 0.6010422698320788, 'overall_recall': 0.6619897959183674, 'overall_f1': 0.6300455235204856, 'overall_accuracy': 0.8370883435898756}
			------------EPOCH 23---------------
Loss:  tensor(0.0210, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1137, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1768, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0634, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0259, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0001, 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(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0270, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2188, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0484, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0556, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0509, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0169, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0024, 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(0.0764, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0940, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0166, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4878048780487805, 'recall': 0.5811965811965812, 'f1': 0.5304212168486739, 'number': 585}, 'P': {'precision': 0.6793002915451894, 'recall': 0.711088504577823, 'f1': 0.6948310139165009, 'number': 983}, 'overall_precision': 0.6019698725376593, 'overall_recall': 0.6626275510204082, 'overall_f1': 0.6308439587128112, 'overall_accuracy': 0.8333805391765584}
			------------EPOCH 24---------------
Loss:  tensor(0.0185, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0165, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1411, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2196, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0734, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0373, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0038, 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(0.0190, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2173, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0187, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0551, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1988, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0630, 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(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0261, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0173, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0968, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1139, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0263, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4984848484848485, 'recall': 0.5623931623931624, 'f1': 0.5285140562248996, 'number': 585}, 'P': {'precision': 0.6623012160898035, 'recall': 0.7202441505595117, 'f1': 0.6900584795321637, 'number': 983}, 'overall_precision': 0.5997686524002314, 'overall_recall': 0.6613520408163265, 'overall_f1': 0.6290567182286927, 'overall_accuracy': 0.8356721682931225}
			------------EPOCH 25---------------
Loss:  tensor(0.0177, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1223, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1943, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0822, 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>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0169, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0129, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0283, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0507, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0777, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0246, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0756, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0865, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0191, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5332197614991482, 'recall': 0.535042735042735, 'f1': 0.5341296928327645, 'number': 585}, 'P': {'precision': 0.6562778272484416, 'recall': 0.7497456765005086, 'f1': 0.6999050332383665, 'number': 983}, 'overall_precision': 0.6140350877192983, 'overall_recall': 0.6696428571428571, 'overall_f1': 0.6406345332519829, 'overall_accuracy': 0.8369081030975616}
			------------EPOCH 26---------------
Loss:  tensor(0.0171, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1199, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0084, 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(0.0073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0664, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0258, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0190, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2164, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0305, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, 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(0.1923, 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(0.0505, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, 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(0.0608, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0725, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0146, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5221674876847291, 'recall': 0.5435897435897435, 'f1': 0.5326633165829147, 'number': 585}, 'P': {'precision': 0.6492007104795737, 'recall': 0.7436419125127162, 'f1': 0.6932195353247985, 'number': 983}, 'overall_precision': 0.6046109510086455, 'overall_recall': 0.6690051020408163, 'overall_f1': 0.6351801392673326, 'overall_accuracy': 0.8332002986842444}
			------------EPOCH 27---------------
Loss:  tensor(0.0193, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0112, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0953, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1742, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0066, 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(0.0081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0181, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0025, 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(0.0198, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1493, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0342, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1489, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0591, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0150, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0855, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0824, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0187, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.49698795180722893, 'recall': 0.5641025641025641, 'f1': 0.5284227381905525, 'number': 585}, 'P': {'precision': 0.6580645161290323, 'recall': 0.7263479145473042, 'f1': 0.690522243713733, 'number': 983}, 'overall_precision': 0.5969125214408233, 'overall_recall': 0.6658163265306123, 'overall_f1': 0.6294844739222188, 'overall_accuracy': 0.8347452171897932}
			------------EPOCH 28---------------
Loss:  tensor(0.0174, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0815, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1448, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0064, 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(0.0081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0527, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0214, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0169, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1616, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0223, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0308, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0433, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1578, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0455, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0513, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0487, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0106, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5114155251141552, 'recall': 0.5743589743589743, 'f1': 0.5410628019323671, 'number': 585}, 'P': {'precision': 0.6758555133079848, 'recall': 0.7232960325534079, 'f1': 0.6987714987714987, 'number': 983}, 'overall_precision': 0.6126389701579871, 'overall_recall': 0.6677295918367347, 'overall_f1': 0.6389990845285322, 'overall_accuracy': 0.8362386384118238}
			------------EPOCH 29---------------
Loss:  tensor(0.0156, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1174, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1682, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0066, 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(0.0066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0115, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0021, 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(0.0152, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1184, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0294, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0346, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1107, 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(0.0222, 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(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0152, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0520, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0599, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0226, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5067669172932331, 'recall': 0.576068376068376, 'f1': 0.5392, 'number': 585}, 'P': {'precision': 0.6595149253731343, 'recall': 0.7192268565615463, 'f1': 0.6880778588807785, 'number': 983}, 'overall_precision': 0.6010362694300518, 'overall_recall': 0.6658163265306123, 'overall_f1': 0.6317700453857792, 'overall_accuracy': 0.8383242783943147}
			------------EPOCH 30---------------
Loss:  tensor(0.0160, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0662, 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(0.0063, 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(0.0062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0333, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0194, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, 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(0.0187, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1199, 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(0.0210, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0467, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1362, 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(0.0208, 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(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0069, 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(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0370, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0380, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0131, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5335570469798657, 'recall': 0.5435897435897435, 'f1': 0.5385266723116002, 'number': 585}, 'P': {'precision': 0.6444833625218914, 'recall': 0.7487283825025433, 'f1': 0.6927058823529413, 'number': 983}, 'overall_precision': 0.6064441887226697, 'overall_recall': 0.6721938775510204, 'overall_f1': 0.6376285541439807, 'overall_accuracy': 0.8346164739809975}
	Train size: 50 Test size: 50


		-------------RUN 1-----------
			------------EPOCH 1---------------
Loss:  tensor(4.1577, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5541, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.7071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.8084, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4833, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2641, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2522, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1368, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9802, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0185, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8834, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6494, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4671, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2687, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3564, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4612, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1829, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4797, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4633, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5743, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.27748132337246534, 'recall': 0.22279348757497858, 'f1': 0.247148288973384, 'number': 1167}, 'P': {'precision': 0.40666381888082015, 'recall': 0.4678132678132678, 'f1': 0.43510054844606944, 'number': 2035}, 'overall_precision': 0.36973764490543015, 'overall_recall': 0.37851342910680824, 'overall_f1': 0.3740740740740741, 'overall_accuracy': 0.6997353659926308}
			------------EPOCH 2---------------
Loss:  tensor(2.3740, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3515, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1834, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1812, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9708, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0755, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0992, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9210, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9317, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3391, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9922, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6335, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2495, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7768, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6173, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7543, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8540, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4491, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.28435114503816794, 'recall': 0.25535561268209084, 'f1': 0.2690744920993228, 'number': 1167}, 'P': {'precision': 0.2618343195266272, 'recall': 0.4348894348894349, 'f1': 0.32686980609418287, 'number': 2035}, 'overall_precision': 0.26716350496838304, 'overall_recall': 0.3694565896314803, 'overall_f1': 0.3100917431192661, 'overall_accuracy': 0.7463185484381529}
			------------EPOCH 3---------------
Loss:  tensor(1.9101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2895, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6461, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5916, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6236, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6477, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1485, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5595, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6536, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5181, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4773, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5626, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2515, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5360, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6247, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2980, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1921, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8621, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4915, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0701, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4752, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3537, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4902, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3355, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.32142857142857145, 'recall': 0.4318766066838046, 'f1': 0.3685557586837295, 'number': 1167}, 'P': {'precision': 0.340683572216097, 'recall': 0.45552825552825554, 'f1': 0.38982338099243063, 'number': 2035}, 'overall_precision': 0.33364420610865003, 'overall_recall': 0.44690818238600877, 'overall_f1': 0.38205847016419703, 'overall_accuracy': 0.7687301366220545}
			------------EPOCH 4---------------
Loss:  tensor(1.4502, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2622, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1937, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3560, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8864, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0112, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1871, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2796, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1241, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2634, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1967, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1998, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4845, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9476, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1386, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6462, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3648, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0722, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3623, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2632, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6077, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3333, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3768584356819651, 'recall': 0.4995715509854327, 'f1': 0.42962417096536476, 'number': 1167}, 'P': {'precision': 0.4473183720045645, 'recall': 0.5778869778869778, 'f1': 0.5042881646655232, 'number': 2035}, 'overall_precision': 0.4212164750957854, 'overall_recall': 0.5493441599000625, 'overall_f1': 0.47682298725941985, 'overall_accuracy': 0.8000557124226041}
			------------EPOCH 5---------------
Loss:  tensor(1.0470, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1757, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9704, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9218, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1324, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6782, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7800, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9157, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9277, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0961, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1355, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3975, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0950, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6423, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1954, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0460, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2915, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9779, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2572, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.33598183881952326, 'recall': 0.5072836332476436, 'f1': 0.4042335268009559, 'number': 1167}, 'P': {'precision': 0.432475884244373, 'recall': 0.5287469287469287, 'f1': 0.47579040459871763, 'number': 2035}, 'overall_precision': 0.3924705882352941, 'overall_recall': 0.5209244222361025, 'overall_f1': 0.4476650563607086, 'overall_accuracy': 0.7991820403408587}
			------------EPOCH 6---------------
Loss:  tensor(0.7564, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1294, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8111, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7632, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7721, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9645, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5282, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7726, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8782, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6501, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7488, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7775, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2972, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0627, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4614, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9813, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6284, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8628, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2040, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.38604076590487957, 'recall': 0.5355612682090831, 'f1': 0.4486719310839914, 'number': 1167}, 'P': {'precision': 0.4350943396226415, 'recall': 0.5665847665847665, 'f1': 0.4922091782283885, 'number': 2035}, 'overall_precision': 0.4164909814944952, 'overall_recall': 0.5552779512804498, 'overall_f1': 0.47597376522553875, 'overall_accuracy': 0.8060321359383111}
			------------EPOCH 7---------------
Loss:  tensor(0.4735, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0998, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5465, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4534, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5726, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7494, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3512, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4933, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5343, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6990, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4341, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0674, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6457, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2521, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4484, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0550, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3900, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9382, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0210, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1927, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3908, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7324, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1779, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.40547588005215124, 'recall': 0.5329905741216795, 'f1': 0.46057015920029615, 'number': 1167}, 'P': {'precision': 0.5209731543624161, 'recall': 0.6103194103194103, 'f1': 0.5621181262729125, 'number': 2035}, 'overall_precision': 0.47575293517100564, 'overall_recall': 0.5821361648969394, 'overall_f1': 0.5235955056179775, 'overall_accuracy': 0.807070412905023}
			------------EPOCH 8---------------
Loss:  tensor(0.3049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0850, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3740, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3923, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3316, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4746, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6251, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3554, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3272, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0770, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5102, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1658, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0257, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3423, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0468, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2289, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5999, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0228, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1940, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.41941624365482233, 'recall': 0.5664095972579263, 'f1': 0.4819540648924535, 'number': 1167}, 'P': {'precision': 0.6112099644128114, 'recall': 0.6751842751842752, 'f1': 0.6416063506887695, 'number': 2035}, 'overall_precision': 0.5321652719665272, 'overall_recall': 0.6355402873204248, 'overall_f1': 0.5792769712496442, 'overall_accuracy': 0.8025374476113298}
			------------EPOCH 9---------------
Loss:  tensor(0.3281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3335, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3314, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5925, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2157, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3276, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3774, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2927, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3382, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4415, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4459, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3952, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0367, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2120, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8595, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1465, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2963, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6808, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2002, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4712328767123288, 'recall': 0.442159383033419, 'f1': 0.4562334217506631, 'number': 1167}, 'P': {'precision': 0.5220375722543352, 'recall': 0.7100737100737101, 'f1': 0.601707266291901, 'number': 2035}, 'overall_precision': 0.5076365519026663, 'overall_recall': 0.6124297314178638, 'overall_f1': 0.5551309271054494, 'overall_accuracy': 0.8034997530926725}
			------------EPOCH 10---------------
Loss:  tensor(0.3857, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0998, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3609, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2835, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3203, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4970, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1939, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2277, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2193, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3882, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3771, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0204, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3597, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1242, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2112, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1400, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2144, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4762, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1352, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4669950738916256, 'recall': 0.40616966580976865, 'f1': 0.4344637946837764, 'number': 1167}, 'P': {'precision': 0.5631067961165048, 'recall': 0.741031941031941, 'f1': 0.6399321026946743, 'number': 2035}, 'overall_precision': 0.5366910370972109, 'overall_recall': 0.6189881324172393, 'overall_f1': 0.5749093546047861, 'overall_accuracy': 0.8089190523823392}
			------------EPOCH 11---------------
Loss:  tensor(0.1628, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0737, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2736, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1975, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4630, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1247, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1225, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1404, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2213, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1506, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1168, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2884, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0126, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3598, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0974, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1833, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0241, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6593, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0657, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1334, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3860, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1333, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5818908122503329, 'recall': 0.3744644387317909, 'f1': 0.4556830031282586, 'number': 1167}, 'P': {'precision': 0.5047744484688838, 'recall': 0.7533169533169534, 'f1': 0.6044952681388013, 'number': 2035}, 'overall_precision': 0.5200633579725449, 'overall_recall': 0.6152404747033104, 'overall_f1': 0.5636623748211731, 'overall_accuracy': 0.8070197652481103}
			------------EPOCH 12---------------
Loss:  tensor(0.0927, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0402, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3466, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1714, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3166, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5945, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1509, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0961, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1239, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1761, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1233, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0944, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2369, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0105, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2805, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2380, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1227, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1603, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4264, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1365, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3936335403726708, 'recall': 0.43444730077120824, 'f1': 0.41303462321792256, 'number': 1167}, 'P': {'precision': 0.443701971266288, 'recall': 0.6525798525798526, 'f1': 0.5282418456642801, 'number': 2035}, 'overall_precision': 0.42863816865218407, 'overall_recall': 0.5730793254216114, 'overall_f1': 0.49044500868635577, 'overall_accuracy': 0.8107297061169708}
			------------EPOCH 13---------------
Loss:  tensor(0.0897, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2303, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0863, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1548, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5406, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3898, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3342, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4211, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1499, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2398, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0506, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1931, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0181, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2116, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7141, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2916, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1599, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2974, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4012801575578533, 'recall': 0.6983718937446444, 'f1': 0.5096935584740463, 'number': 1167}, 'P': {'precision': 0.6539923954372624, 'recall': 0.5071253071253071, 'f1': 0.5712704123996679, 'number': 2035}, 'overall_precision': 0.5117761152673871, 'overall_recall': 0.5768269831355403, 'overall_f1': 0.5423579503743944, 'overall_accuracy': 0.7773149144687694}
			------------EPOCH 14---------------
Loss:  tensor(0.3989, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1798, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1535, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0911, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1699, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2107, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4174, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3511, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5126, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0854, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4416, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0860, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0128, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1705, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6550, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0653, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1139, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3240, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1242, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.37664473684210525, 'recall': 0.5886889460154242, 'f1': 0.4593781344032096, 'number': 1167}, 'P': {'precision': 0.5502304147465438, 'recall': 0.5867321867321867, 'f1': 0.5678953626634959, 'number': 2035}, 'overall_precision': 0.47095643465197795, 'overall_recall': 0.5874453466583386, 'overall_f1': 0.5227904391328516, 'overall_accuracy': 0.8044240728313307}
			------------EPOCH 15---------------
Loss:  tensor(0.0999, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2176, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2244, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1706, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0965, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1114, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1496, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1817, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0548, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0847, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4774, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0550, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0967, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2904, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1275, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.44850746268656716, 'recall': 0.5149957155098543, 'f1': 0.47945751894694855, 'number': 1167}, 'P': {'precision': 0.5942152275627393, 'recall': 0.6864864864864865, 'f1': 0.6370269037847698, 'number': 2035}, 'overall_precision': 0.541316716337036, 'overall_recall': 0.6239850093691443, 'overall_f1': 0.5797185550558537, 'overall_accuracy': 0.8187067120807323}
			------------EPOCH 16---------------
Loss:  tensor(0.0523, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1475, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0310, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0470, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2917, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0566, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0451, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0597, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1202, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0858, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0448, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1412, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0105, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0272, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0672, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4187, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0194, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0563, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2166, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1149, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.46023564064801176, 'recall': 0.5355612682090831, 'f1': 0.49504950495049505, 'number': 1167}, 'P': {'precision': 0.6415180612711477, 'recall': 0.6894348894348894, 'f1': 0.6646139270487921, 'number': 2035}, 'overall_precision': 0.5720733427362482, 'overall_recall': 0.6333541536539663, 'overall_f1': 0.6011560693641619, 'overall_accuracy': 0.8187700216518733}
			------------EPOCH 17---------------
Loss:  tensor(0.0343, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0308, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2763, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0407, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0255, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0414, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0779, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1188, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0168, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0401, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4111, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0453, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1982, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1057, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4107965766951942, 'recall': 0.5347043701799485, 'f1': 0.4646314221891288, 'number': 1167}, 'P': {'precision': 0.5748373101952278, 'recall': 0.6511056511056511, 'f1': 0.6105990783410139, 'number': 2035}, 'overall_precision': 0.5096757322175732, 'overall_recall': 0.608682073703935, 'overall_f1': 0.5547964702533447, 'overall_accuracy': 0.8142750421008648}
			------------EPOCH 18---------------
Loss:  tensor(0.0194, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0966, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0186, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0185, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0488, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0167, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0246, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0508, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0175, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0808, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0424, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3943, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0428, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1618, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0939, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.44667143879742305, 'recall': 0.5347043701799485, 'f1': 0.4867394695787831, 'number': 1167}, 'P': {'precision': 0.6156914893617021, 'recall': 0.6825552825552825, 'f1': 0.6474015381030063, 'number': 2035}, 'overall_precision': 0.5510539282781276, 'overall_recall': 0.6286695815115553, 'overall_f1': 0.5873085339168489, 'overall_accuracy': 0.8163895817769731}
			------------EPOCH 19---------------
Loss:  tensor(0.0155, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0770, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0127, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0136, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0402, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0116, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0174, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0335, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0163, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0681, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0900, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0316, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3513, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0372, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1402, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0829, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4474982381959126, 'recall': 0.5441302485004285, 'f1': 0.4911059551430781, 'number': 1167}, 'P': {'precision': 0.6156914893617021, 'recall': 0.6825552825552825, 'f1': 0.6474015381030063, 'number': 2035}, 'overall_precision': 0.5507482993197279, 'overall_recall': 0.63210493441599, 'overall_f1': 0.588628762541806, 'overall_accuracy': 0.8174911683148258}
			------------EPOCH 20---------------
Loss:  tensor(0.0141, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0675, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0114, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0140, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1969, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0593, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0160, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0244, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0555, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0791, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0228, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3390, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0292, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1167, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0792, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.42467620995228356, 'recall': 0.5338474721508141, 'f1': 0.4730447987851177, 'number': 1167}, 'P': {'precision': 0.5957074025405169, 'recall': 0.6683046683046683, 'f1': 0.6299212598425197, 'number': 2035}, 'overall_precision': 0.5288, 'overall_recall': 0.6193004372267333, 'overall_f1': 0.5704833141542003, 'overall_accuracy': 0.8154272762956304}
			------------EPOCH 21---------------
Loss:  tensor(0.0126, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0621, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0102, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1660, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0538, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0133, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0107, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0476, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0716, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3128, 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(0.0264, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1000, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0749, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4554819720382634, 'recall': 0.5304198800342759, 'f1': 0.49010292953285817, 'number': 1167}, 'P': {'precision': 0.6038227628149435, 'recall': 0.683046683046683, 'f1': 0.640996080239797, 'number': 2035}, 'overall_precision': 0.5487571701720841, 'overall_recall': 0.627420362273579, 'overall_f1': 0.5854582544076934, 'overall_accuracy': 0.8171872823733493}
			------------EPOCH 22---------------
Loss:  tensor(0.0121, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0592, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1614, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0334, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0077, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0143, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0435, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0624, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0164, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2921, 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(0.0235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0989, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0744, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.438385269121813, 'recall': 0.5304198800342759, 'f1': 0.4800310197751066, 'number': 1167}, 'P': {'precision': 0.6043324491600354, 'recall': 0.6717444717444717, 'f1': 0.6362578543169654, 'number': 2035}, 'overall_precision': 0.5405552531301034, 'overall_recall': 0.6202373516552155, 'overall_f1': 0.5776614310645725, 'overall_accuracy': 0.8152373475822075}
			------------EPOCH 23---------------
Loss:  tensor(0.0111, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0624, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1429, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0467, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0391, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0562, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0184, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2824, 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(0.0208, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0886, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0677, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.42692046227056424, 'recall': 0.5381319622964867, 'f1': 0.4761182714177407, 'number': 1167}, 'P': {'precision': 0.6051123843102688, 'recall': 0.6746928746928746, 'f1': 0.6380111524163568, 'number': 2035}, 'overall_precision': 0.5350267379679144, 'overall_recall': 0.6249219237976265, 'overall_f1': 0.5764909248055314, 'overall_accuracy': 0.8162882864631475}
			------------EPOCH 24---------------
Loss:  tensor(0.0106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0453, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1307, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0547, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0087, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0441, 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(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2737, 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(0.0169, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0669, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0673, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.44204463642908565, 'recall': 0.5261353898886033, 'f1': 0.4804381846635368, 'number': 1167}, 'P': {'precision': 0.6073619631901841, 'recall': 0.6810810810810811, 'f1': 0.6421125781792911, 'number': 2035}, 'overall_precision': 0.5448106782892945, 'overall_recall': 0.6246096189881324, 'overall_f1': 0.5819874872690238, 'overall_accuracy': 0.8173518872583158}
			------------EPOCH 25---------------
Loss:  tensor(0.0102, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0410, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0520, 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(0.0077, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0324, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0361, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0029, 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(0.2514, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0121, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0651, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0667, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4375446109921485, 'recall': 0.5252784918594687, 'f1': 0.47741433021806856, 'number': 1167}, 'P': {'precision': 0.6021834061135372, 'recall': 0.6776412776412777, 'f1': 0.6376878612716764, 'number': 2035}, 'overall_precision': 0.5396911406123002, 'overall_recall': 0.6221111805121798, 'overall_f1': 0.5779776584941244, 'overall_accuracy': 0.8158451194651607}
			------------EPOCH 26---------------
Loss:  tensor(0.0088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0351, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0888, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0332, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0064, 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(0.0061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0288, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0282, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2423, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0567, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0575, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.42926155969634233, 'recall': 0.5329905741216795, 'f1': 0.47553516819571867, 'number': 1167}, 'P': {'precision': 0.6203496190049306, 'recall': 0.6800982800982801, 'f1': 0.6488513830285982, 'number': 2035}, 'overall_precision': 0.5451086956521739, 'overall_recall': 0.6264834478450968, 'overall_f1': 0.5829700668410346, 'overall_accuracy': 0.8160350481785836}
			------------EPOCH 27---------------
Loss:  tensor(0.0075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0352, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0987, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0288, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0060, 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(0.0255, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0260, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2432, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, 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(0.0627, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0611, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.48324240062353857, 'recall': 0.5312767780634104, 'f1': 0.5061224489795918, 'number': 1167}, 'P': {'precision': 0.6161879895561357, 'recall': 0.6958230958230959, 'f1': 0.6535887375951995, 'number': 2035}, 'overall_precision': 0.5685562691985478, 'overall_recall': 0.6358525921299188, 'overall_f1': 0.6003243402624208, 'overall_accuracy': 0.8178710257416716}
			------------EPOCH 28---------------
Loss:  tensor(0.0078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0439, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1257, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0271, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0425, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0676, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2759, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0659, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4057262569832402, 'recall': 0.49785775492716366, 'f1': 0.4470950365525202, 'number': 1167}, 'P': {'precision': 0.5734858110969928, 'recall': 0.6653562653562654, 'f1': 0.6160145586897179, 'number': 2035}, 'overall_precision': 0.5101502768257317, 'overall_recall': 0.6043098063710182, 'overall_f1': 0.55325232308792, 'overall_accuracy': 0.8130468364207301}
			------------EPOCH 29---------------
Loss:  tensor(0.0075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0662, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0085, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0265, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0499, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0701, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0234, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3504, 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(0.0137, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1410, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0632, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4449826989619377, 'recall': 0.5509854327335048, 'f1': 0.4923430321592649, 'number': 1167}, 'P': {'precision': 0.6117232260907889, 'recall': 0.682063882063882, 'f1': 0.6449814126394051, 'number': 2035}, 'overall_precision': 0.5468497576736672, 'overall_recall': 0.6342910680824485, 'overall_f1': 0.5873337189126663, 'overall_accuracy': 0.8152500094964357}
			------------EPOCH 30---------------
Loss:  tensor(0.0081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0447, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1975, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0265, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0457, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0643, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0036, 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(0.2710, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0639, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0522, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4700149925037481, 'recall': 0.5372750642673522, 'f1': 0.5013994402239105, 'number': 1167}, 'P': {'precision': 0.6206743566992015, 'recall': 0.6874692874692875, 'f1': 0.6523665190020983, 'number': 2035}, 'overall_precision': 0.564659977703456, 'overall_recall': 0.6327295440349782, 'overall_f1': 0.5967599410898381, 'overall_accuracy': 0.8156931764944224}


		-------------RUN 2-----------
			------------EPOCH 1---------------
Loss:  tensor(2.4756, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8505, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2364, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2333, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8630, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7715, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7511, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6811, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8506, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4864, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1165, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5925, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3912, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2894, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2753, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1930, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2776, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9225, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4525, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9677, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2884, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2845528455284553, 'recall': 0.09408602150537634, 'f1': 0.1414141414141414, 'number': 1116}, 'P': {'precision': 0.45320197044334976, 'recall': 0.6013071895424836, 'f1': 0.5168539325842696, 'number': 1836}, 'overall_precision': 0.4310160427807487, 'overall_recall': 0.4095528455284553, 'overall_f1': 0.420010422094841, 'overall_accuracy': 0.7184185279770833}
			------------EPOCH 2---------------
Loss:  tensor(1.4187, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6084, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2632, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3736, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8824, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8200, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6595, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7724, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7845, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7167, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1164, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5447, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5952, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4163, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6473, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7381, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6225, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6622, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3768, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3911, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2221, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.375886524822695, 'recall': 0.3324372759856631, 'f1': 0.35282929148834996, 'number': 1116}, 'P': {'precision': 0.4795505617977528, 'recall': 0.5811546840958606, 'f1': 0.5254863334154148, 'number': 1836}, 'overall_precision': 0.4476961394769614, 'overall_recall': 0.48712737127371275, 'overall_f1': 0.4665801427644387, 'overall_accuracy': 0.7736160092964179}
			------------EPOCH 3---------------
Loss:  tensor(1.0443, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2467, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9579, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4372, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2459, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3422, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3804, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8673, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1652, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2581, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1976, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2267, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2691, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3207, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1538, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2841, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2506, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4835, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2827, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0685, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1636, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4002659574468085, 'recall': 0.5394265232974911, 'f1': 0.4595419847328244, 'number': 1116}, 'P': {'precision': 0.5583677685950413, 'recall': 0.5887799564270153, 'f1': 0.5731707317073171, 'number': 1836}, 'overall_precision': 0.48924418604651165, 'overall_recall': 0.5701219512195121, 'overall_f1': 0.5265957446808511, 'overall_accuracy': 0.7934114340535354}
			------------EPOCH 4---------------
Loss:  tensor(0.7816, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9930, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7303, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8454, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1364, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1969, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9716, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1202, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0400, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9787, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6441, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8791, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0857, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2714, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9193, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1803, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1309, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8802, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8605, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0966, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3434, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2153, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7667, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1277, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.40180296200901483, 'recall': 0.5591397849462365, 'f1': 0.4675908579992506, 'number': 1116}, 'P': {'precision': 0.6076712328767123, 'recall': 0.6040305010893247, 'f1': 0.6058453974323955, 'number': 1836}, 'overall_precision': 0.5130254588513914, 'overall_recall': 0.587059620596206, 'overall_f1': 0.5475513428120063, 'overall_accuracy': 0.8104773872741767}
			------------EPOCH 5---------------
Loss:  tensor(0.5507, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5513, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6826, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8343, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9726, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7378, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9257, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6955, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4915, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8719, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2294, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6663, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7834, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6321, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5658, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2313, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1704, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0868, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.43428912783751494, 'recall': 0.6514336917562724, 'f1': 0.5211469534050179, 'number': 1116}, 'P': {'precision': 0.6321709786276716, 'recall': 0.6122004357298475, 'f1': 0.6220254565578307, 'number': 1836}, 'overall_precision': 0.5362108922363847, 'overall_recall': 0.6270325203252033, 'overall_f1': 0.5780762023735165, 'overall_accuracy': 0.806477765616766}
			------------EPOCH 6---------------
Loss:  tensor(0.4174, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5996, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3895, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5692, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5790, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7671, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4629, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4589, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3288, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4621, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7894, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5098, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5280, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8738, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3894, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3626, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8442, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1435, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4718, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0566, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4368279569892473, 'recall': 0.5824372759856631, 'f1': 0.49923195084485417, 'number': 1116}, 'P': {'precision': 0.6031168831168832, 'recall': 0.6323529411764706, 'f1': 0.6173889922892848, 'number': 1836}, 'overall_precision': 0.5306182244359801, 'overall_recall': 0.6134823848238482, 'overall_f1': 0.5690494893951296, 'overall_accuracy': 0.8180307268231384}
			------------EPOCH 7---------------
Loss:  tensor(0.3056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3934, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2638, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4727, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5228, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7563, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3662, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1956, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3177, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6948, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1809, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3740, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0841, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3671, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6242, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3524, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1649, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1116, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2888, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0495, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.32246998284734135, 'recall': 0.6738351254480287, 'f1': 0.43619489559164737, 'number': 1116}, 'P': {'precision': 0.58, 'recall': 0.4106753812636166, 'f1': 0.48086734693877553, 'number': 1836}, 'overall_precision': 0.414647577092511, 'overall_recall': 0.5101626016260162, 'overall_f1': 0.45747266099635475, 'overall_accuracy': 0.7357277014336482}
			------------EPOCH 8---------------
Loss:  tensor(0.3373, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4901, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2657, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5923, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4164, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6368, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6874, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4998, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1562, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2646, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1495, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3933, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0681, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5679, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7931, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3350, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5997, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1305, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1423, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3947, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0772, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5193298969072165, 'recall': 0.3611111111111111, 'f1': 0.4260042283298097, 'number': 1116}, 'P': {'precision': 0.5395161290322581, 'recall': 0.7287581699346405, 'f1': 0.6200185356811863, 'number': 1836}, 'overall_precision': 0.5347051597051597, 'overall_recall': 0.589769647696477, 'overall_f1': 0.560889175257732, 'overall_accuracy': 0.800951261367168}
			------------EPOCH 9---------------
Loss:  tensor(0.3917, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3967, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4712, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3770, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6241, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2909, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4775, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2333, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2607, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1757, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2517, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7622, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2211, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2362, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0436, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2663, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1521, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1909, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6720, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0266, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5191570881226054, 'recall': 0.48566308243727596, 'f1': 0.5018518518518519, 'number': 1116}, 'P': {'precision': 0.5972354623450906, 'recall': 0.6824618736383442, 'f1': 0.6370106761565836, 'number': 1836}, 'overall_precision': 0.5712921705919797, 'overall_recall': 0.6080623306233063, 'overall_f1': 0.5891040367574665, 'overall_accuracy': 0.8220979096571945}
			------------EPOCH 10---------------
Loss:  tensor(0.2136, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1840, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1191, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3351, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2715, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4755, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1432, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4306, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1600, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0493, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0813, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4637, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1441, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0251, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1573, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3714, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0824, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0881, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5196, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0922, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0826, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0809, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0076, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4260929909784872, 'recall': 0.5501792114695341, 'f1': 0.4802502933124756, 'number': 1116}, 'P': {'precision': 0.5506208213944603, 'recall': 0.6279956427015251, 'f1': 0.5867684478371501, 'number': 1836}, 'overall_precision': 0.49985855728429984, 'overall_recall': 0.5985772357723578, 'overall_f1': 0.5447818714351781, 'overall_accuracy': 0.8214763468320564}
			------------EPOCH 11---------------
Loss:  tensor(0.1000, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0901, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0704, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2868, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2102, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3814, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0797, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3421, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0488, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0877, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0557, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0770, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5313, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1342, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0259, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1120, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3476, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0490, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4809, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0684, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0664, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0616, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0072, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.38277202072538863, 'recall': 0.5295698924731183, 'f1': 0.4443609022556391, 'number': 1116}, 'P': {'precision': 0.4688372093023256, 'recall': 0.5490196078431373, 'f1': 0.5057701956848971, 'number': 1836}, 'overall_precision': 0.4328641039523552, 'overall_recall': 0.5416666666666666, 'overall_f1': 0.4811916942521817, 'overall_accuracy': 0.810072020214304}
			------------EPOCH 12---------------
Loss:  tensor(0.0812, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0937, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0859, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2662, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2199, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3430, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0632, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3241, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0343, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0697, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0379, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3980, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0130, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1873, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3724, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0544, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0789, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4213, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0627, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0571, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0362, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0051, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4215817694369973, 'recall': 0.5636200716845878, 'f1': 0.48236196319018404, 'number': 1116}, 'P': {'precision': 0.50523311132255, 'recall': 0.5784313725490197, 'f1': 0.5393600812595227, 'number': 1836}, 'overall_precision': 0.4705063995548136, 'overall_recall': 0.5728319783197832, 'overall_f1': 0.5166513901619308, 'overall_accuracy': 0.8159228181118002}
			------------EPOCH 13---------------
Loss:  tensor(0.0662, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0603, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0783, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2420, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3562, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1324, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3514, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0745, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1207, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0381, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0587, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3700, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0994, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0564, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0112, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0923, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2742, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0542, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0727, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5307, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1077, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1346, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0116, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5127118644067796, 'recall': 0.5421146953405018, 'f1': 0.5270034843205575, 'number': 1116}, 'P': {'precision': 0.6590678824721378, 'recall': 0.7086056644880174, 'f1': 0.6829396325459317, 'number': 1836}, 'overall_precision': 0.6043119847812302, 'overall_recall': 0.6456639566395664, 'overall_f1': 0.6243039633147724, 'overall_accuracy': 0.8307322280324834}
			------------EPOCH 14---------------
Loss:  tensor(0.0820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0828, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0439, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2621, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1319, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2863, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0356, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2626, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0797, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0826, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0518, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0963, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5493, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0596, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0148, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0729, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2642, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0359, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0337, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4308, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0596, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0626, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0571, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0162, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.46531566640685895, 'recall': 0.5349462365591398, 'f1': 0.4977073780741976, 'number': 1116}, 'P': {'precision': 0.5775635407537248, 'recall': 0.7178649237472767, 'f1': 0.6401165614375912, 'number': 1836}, 'overall_precision': 0.5371669004207573, 'overall_recall': 0.6487127371273713, 'overall_f1': 0.5876937241061838, 'overall_accuracy': 0.814841839285473}
			------------EPOCH 15---------------
Loss:  tensor(0.1933, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1196, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0646, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3254, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0653, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3357, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0625, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0643, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0166, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0310, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0867, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0905, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0515, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1137, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2455, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0447, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4372, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0682, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0385, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0027, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4727653631284916, 'recall': 0.6066308243727598, 'f1': 0.5313971742543171, 'number': 1116}, 'P': {'precision': 0.6123061530765382, 'recall': 0.6666666666666666, 'f1': 0.6383311603650587, 'number': 1836}, 'overall_precision': 0.5540658700087439, 'overall_recall': 0.643970189701897, 'overall_f1': 0.5956446811843961, 'overall_accuracy': 0.8262731903738836}
			------------EPOCH 16---------------
Loss:  tensor(0.0493, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0296, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0323, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1369, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2857, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0503, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0214, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0429, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0143, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0223, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0852, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0412, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1793, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0154, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0339, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3757, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0394, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0497, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0185, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0025, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4902597402597403, 'recall': 0.5412186379928315, 'f1': 0.514480408858603, 'number': 1116}, 'P': {'precision': 0.6411411411411412, 'recall': 0.6977124183006536, 'f1': 0.6682316118935838, 'number': 1836}, 'overall_precision': 0.5835913312693498, 'overall_recall': 0.638550135501355, 'overall_f1': 0.6098350048527983, 'overall_accuracy': 0.830070128501358}
			------------EPOCH 17---------------
Loss:  tensor(0.0415, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0271, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0195, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2526, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2202, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0121, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0350, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0098, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0181, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4180, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1168, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0193, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0342, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1733, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0150, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3616, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0386, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0440, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0150, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4694736842105263, 'recall': 0.5994623655913979, 'f1': 0.5265643447461629, 'number': 1116}, 'P': {'precision': 0.6489994591671173, 'recall': 0.6535947712418301, 'f1': 0.6512890094979646, 'number': 1836}, 'overall_precision': 0.5708613317043372, 'overall_recall': 0.633130081300813, 'overall_f1': 0.6003854802441375, 'overall_accuracy': 0.8263542637858581}
			------------EPOCH 18---------------
Loss:  tensor(0.0480, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0141, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0168, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1810, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2464, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2204, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0085, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0284, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0112, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3276, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0764, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0107, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0266, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1329, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0130, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3753, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0409, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0518, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0135, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5053859964093357, 'recall': 0.5044802867383512, 'f1': 0.504932735426009, 'number': 1116}, 'P': {'precision': 0.5861607142857143, 'recall': 0.7151416122004357, 'f1': 0.6442590775269873, 'number': 1836}, 'overall_precision': 0.5593321407274896, 'overall_recall': 0.6355013550135501, 'overall_f1': 0.594988899460831, 'overall_accuracy': 0.8269488021403381}
			------------EPOCH 19---------------
Loss:  tensor(0.0777, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0105, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1896, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0938, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1669, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0204, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3546, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0924, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0102, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0231, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1463, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0114, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3411, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0297, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0370, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4864864864864865, 'recall': 0.5483870967741935, 'f1': 0.5155855096882899, 'number': 1116}, 'P': {'precision': 0.6240893637688198, 'recall': 0.6998910675381264, 'f1': 0.6598202824133504, 'number': 1836}, 'overall_precision': 0.5719023213747362, 'overall_recall': 0.6426151761517616, 'overall_f1': 0.605200191418089, 'overall_accuracy': 0.8294891023822071}
			------------EPOCH 20---------------
Loss:  tensor(0.0324, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1379, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0628, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1823, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1540, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0167, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2552, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0662, 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(0.0174, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0956, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2778, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0173, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0315, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.47976011994003, 'recall': 0.5734767025089605, 'f1': 0.5224489795918368, 'number': 1116}, 'P': {'precision': 0.6463917525773196, 'recall': 0.6830065359477124, 'f1': 0.6641949152542372, 'number': 1836}, 'overall_precision': 0.5784972510690287, 'overall_recall': 0.6415989159891599, 'overall_f1': 0.6084163186636685, 'overall_accuracy': 0.8290972475576635}
			------------EPOCH 21---------------
Loss:  tensor(0.0463, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1212, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0467, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1611, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1230, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0115, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2255, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0593, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0800, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0084, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2851, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0274, 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>)
				 {'C': {'precision': 0.48668280871670705, 'recall': 0.5403225806451613, 'f1': 0.5121019108280255, 'number': 1116}, 'P': {'precision': 0.6288812222769837, 'recall': 0.6949891067538126, 'f1': 0.6602846054333765, 'number': 1836}, 'overall_precision': 0.574969400244798, 'overall_recall': 0.6365176151761518, 'overall_f1': 0.6041800643086817, 'overall_accuracy': 0.8298133960301053}
			------------EPOCH 22---------------
Loss:  tensor(0.0346, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0997, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0435, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1339, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2341, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0641, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0105, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0663, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2432, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0241, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.47495961227786754, 'recall': 0.5268817204301075, 'f1': 0.49957519116397614, 'number': 1116}, 'P': {'precision': 0.5981176470588235, 'recall': 0.6922657952069716, 'f1': 0.6417571320373643, 'number': 1836}, 'overall_precision': 0.5527802557240559, 'overall_recall': 0.6297425474254743, 'overall_f1': 0.5887569279493271, 'overall_accuracy': 0.8286648560271326}
			------------EPOCH 23---------------
Loss:  tensor(0.0291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0972, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0448, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1648, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0993, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1988, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0549, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0631, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2884, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0246, 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>)
				 {'C': {'precision': 0.48564593301435405, 'recall': 0.5456989247311828, 'f1': 0.5139240506329112, 'number': 1116}, 'P': {'precision': 0.6296479920674268, 'recall': 0.6917211328976035, 'f1': 0.6592265766934856, 'number': 1836}, 'overall_precision': 0.5744420666462855, 'overall_recall': 0.6365176151761518, 'overall_f1': 0.603888799614334, 'overall_accuracy': 0.8288134906157526}
			------------EPOCH 24---------------
Loss:  tensor(0.0263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0853, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0308, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0849, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0664, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0645, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0214, 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>)
				 {'C': {'precision': 0.45348837209302323, 'recall': 0.5591397849462365, 'f1': 0.5008025682182986, 'number': 1116}, 'P': {'precision': 0.6407061266874351, 'recall': 0.6721132897603486, 'f1': 0.6560340244550771, 'number': 1836}, 'overall_precision': 0.5626892792247123, 'overall_recall': 0.6294037940379403, 'overall_f1': 0.5941797249760155, 'overall_accuracy': 0.8274082181415272}
			------------EPOCH 25---------------
Loss:  tensor(0.0304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1133, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0411, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1747, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0985, 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(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1772, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0077, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0851, 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(0.2868, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0200, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0294, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.44749216300940436, 'recall': 0.5116487455197133, 'f1': 0.4774247491638796, 'number': 1116}, 'P': {'precision': 0.6, 'recall': 0.684640522875817, 'f1': 0.6395319257186467, 'number': 1836}, 'overall_precision': 0.5422723227528923, 'overall_recall': 0.6192411924119241, 'overall_f1': 0.5782065475249091, 'overall_accuracy': 0.82662450849244}
			------------EPOCH 26---------------
Loss:  tensor(0.0362, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1474, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0401, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1381, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0980, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2741, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0707, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0077, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1310, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2643, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0148, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4856439704675964, 'recall': 0.5304659498207885, 'f1': 0.5070663811563169, 'number': 1116}, 'P': {'precision': 0.6008364312267658, 'recall': 0.704248366013072, 'f1': 0.6484453360080241, 'number': 1836}, 'overall_precision': 0.5591812518540492, 'overall_recall': 0.638550135501355, 'overall_f1': 0.5962359639411671, 'overall_accuracy': 0.8278000729660708}
			------------EPOCH 27---------------
Loss:  tensor(0.0284, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1351, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0497, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1521, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0913, 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(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0539, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0748, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2537, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0137, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0247, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.48822605965463106, 'recall': 0.557347670250896, 'f1': 0.5205020920502091, 'number': 1116}, 'P': {'precision': 0.631815907953977, 'recall': 0.6879084967320261, 'f1': 0.658670143415906, 'number': 1836}, 'overall_precision': 0.5759242285365108, 'overall_recall': 0.638550135501355, 'overall_f1': 0.6056224899598394, 'overall_accuracy': 0.8292053454402962}
			------------EPOCH 28---------------
Loss:  tensor(0.0257, 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(0.0998, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0273, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0565, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1623, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0526, 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(0.0057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0490, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2234, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0085, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0180, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4537313432835821, 'recall': 0.5448028673835126, 'f1': 0.49511400651465803, 'number': 1116}, 'P': {'precision': 0.6238805970149254, 'recall': 0.6830065359477124, 'f1': 0.6521060842433697, 'number': 1836}, 'overall_precision': 0.5558208955223881, 'overall_recall': 0.6307588075880759, 'overall_f1': 0.5909235163440177, 'overall_accuracy': 0.8304619833259016}
			------------EPOCH 29---------------
Loss:  tensor(0.0275, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0264, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0468, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0025, 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(0.1598, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0493, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1918, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0144, 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>)
				 {'C': {'precision': 0.5004240882103478, 'recall': 0.5286738351254481, 'f1': 0.514161220043573, 'number': 1116}, 'P': {'precision': 0.6147657359204922, 'recall': 0.7075163398692811, 'f1': 0.6578880729298557, 'number': 1836}, 'overall_precision': 0.5738153098420413, 'overall_recall': 0.6399051490514905, 'overall_f1': 0.6050608584240871, 'overall_accuracy': 0.8287189049684489}
			------------EPOCH 30---------------
Loss:  tensor(0.0253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0682, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0246, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0876, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0424, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, 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(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1275, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0222, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1980, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0140, 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>)
				 {'C': {'precision': 0.4796812749003984, 'recall': 0.5394265232974911, 'f1': 0.5078026149304091, 'number': 1116}, 'P': {'precision': 0.6113789778206364, 'recall': 0.690631808278867, 'f1': 0.6485933503836318, 'number': 1836}, 'overall_precision': 0.5617302493241214, 'overall_recall': 0.6334688346883469, 'overall_f1': 0.5954465849387042, 'overall_accuracy': 0.8292188576756253}


		-------------RUN 3-----------
			------------EPOCH 1---------------
Loss:  tensor(3.2922, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1409, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9555, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0490, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5653, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1835, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1831, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0147, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2955, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4909, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1712, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3213, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2891, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5407, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1326, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0947, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1639, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9233, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0656, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4342, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2895, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8746, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4223, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1480, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 1119}, 'P': {'precision': 0.42392663609837433, 'recall': 0.5441412520064205, 'f1': 0.4765698219306466, 'number': 1869}, 'overall_precision': 0.423044925124792, 'overall_recall': 0.34036144578313254, 'overall_f1': 0.3772255192878338, 'overall_accuracy': 0.6975951524332513}
			------------EPOCH 2---------------
Loss:  tensor(1.8479, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8266, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8520, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6751, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7451, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6314, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5954, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6910, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4978, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5613, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2087, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0290, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4570, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4562, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8587, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3485, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4788, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9622, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9473, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2698, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3126, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1161, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.24684431977559607, 'recall': 0.1572832886505809, 'f1': 0.19213973799126638, 'number': 1119}, 'P': {'precision': 0.5004226542688082, 'recall': 0.6334938469769931, 'f1': 0.5591499409681228, 'number': 1869}, 'overall_precision': 0.44170185125040595, 'overall_recall': 0.45515394912985274, 'overall_f1': 0.4483270149991759, 'overall_accuracy': 0.7403089241756161}
			------------EPOCH 3---------------
Loss:  tensor(1.3580, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4414, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1603, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2748, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3524, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2573, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0690, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0573, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1466, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0849, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7887, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0429, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0846, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6633, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0098, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0437, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6824, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8853, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2273, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0713, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.37737226277372266, 'recall': 0.46201966041108133, 'f1': 0.4154278826838088, 'number': 1119}, 'P': {'precision': 0.5413064361191162, 'recall': 0.602996254681648, 'f1': 0.5704884839281195, 'number': 1869}, 'overall_precision': 0.4762456546929316, 'overall_recall': 0.5502008032128514, 'overall_f1': 0.5105590062111801, 'overall_accuracy': 0.7974680120107123}
			------------EPOCH 4---------------
Loss:  tensor(1.0596, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1328, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9116, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9498, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9653, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3479, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0894, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3609, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8647, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9509, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7898, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0982, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8465, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8870, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7847, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5132, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9300, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8391, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5482, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4955, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6544, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0629, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.39615384615384613, 'recall': 0.3681858802502234, 'f1': 0.38165817508105604, 'number': 1119}, 'P': {'precision': 0.5489309210526315, 'recall': 0.7142857142857143, 'f1': 0.6207858637526156, 'number': 1869}, 'overall_precision': 0.503168202764977, 'overall_recall': 0.5846720214190094, 'overall_f1': 0.5408668730650156, 'overall_accuracy': 0.8019990802607732}
			------------EPOCH 5---------------
Loss:  tensor(0.8697, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9402, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7643, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7272, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7678, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2647, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9529, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2856, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6536, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6788, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6234, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6502, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0651, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6184, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5562, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5482, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4142, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6424, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6430, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3766, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5414, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1356, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0356, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3875072296124928, 'recall': 0.5987488829311886, 'f1': 0.47050561797752816, 'number': 1119}, 'P': {'precision': 0.6255707762557078, 'recall': 0.5864098448368111, 'f1': 0.6053576360121514, 'number': 1869}, 'overall_precision': 0.5073254811835679, 'overall_recall': 0.5910307898259706, 'overall_f1': 0.545988560828567, 'overall_accuracy': 0.8080450131198096}
			------------EPOCH 6---------------
Loss:  tensor(0.7550, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8105, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5321, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5776, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1806, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9410, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2721, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5488, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6586, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4377, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4600, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0488, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4605, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5924, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5499, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5184, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3255, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5662, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4507, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2801, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2592, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4409, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0337, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3970917225950783, 'recall': 0.31724754244861486, 'f1': 0.35270740188772975, 'number': 1119}, 'P': {'precision': 0.549553208773355, 'recall': 0.723916532905297, 'f1': 0.624797968136689, 'number': 1869}, 'overall_precision': 0.5089392133492253, 'overall_recall': 0.571619812583668, 'overall_f1': 0.5384615384615385, 'overall_accuracy': 0.7901371493494197}
			------------EPOCH 7---------------
Loss:  tensor(0.7302, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7451, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4494, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4140, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4741, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1554, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7933, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1394, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3630, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3636, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4148, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0315, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2978, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4541, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3629, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3844, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2615, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3168, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1987, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2737, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1085, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0185, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.45747316267547483, 'recall': 0.49508489722966936, 'f1': 0.4755364806866953, 'number': 1119}, 'P': {'precision': 0.6292022077270446, 'recall': 0.6709470304975923, 'f1': 0.649404453650958, 'number': 1869}, 'overall_precision': 0.5642946317103621, 'overall_recall': 0.6050870147255689, 'overall_f1': 0.5839793281653748, 'overall_accuracy': 0.8137933832877973}
			------------EPOCH 8---------------
Loss:  tensor(0.4018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5645, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2467, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2220, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3657, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5921, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0914, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2685, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2471, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2497, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2814, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0277, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1927, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3632, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2824, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2616, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1885, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3618, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1931, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1612, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1511, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2556, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1236, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0144, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5223880597014925, 'recall': 0.43789097408400357, 'f1': 0.47642197374817696, 'number': 1119}, 'P': {'precision': 0.6392737697085523, 'recall': 0.7158908507223114, 'f1': 0.6754164563351842, 'number': 1869}, 'overall_precision': 0.603101286704058, 'overall_recall': 0.6117804551539491, 'overall_f1': 0.6074098687489616, 'overall_accuracy': 0.8185814374983093}
			------------EPOCH 9---------------
Loss:  tensor(0.3408, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4302, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2193, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1908, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2484, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5148, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0580, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2259, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1770, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2406, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1234, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2572, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2301, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1702, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3120, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1362, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1411, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1764, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0970, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0149, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5116033755274262, 'recall': 0.4334226988382484, 'f1': 0.4692791485244315, 'number': 1119}, 'P': {'precision': 0.6225619399051133, 'recall': 0.631888710540396, 'f1': 0.6271906532129581, 'number': 1869}, 'overall_precision': 0.5855887521968366, 'overall_recall': 0.5575635876840697, 'overall_f1': 0.5712326418652495, 'overall_accuracy': 0.7872020991695295}
			------------EPOCH 10---------------
Loss:  tensor(0.2772, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5525, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1353, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2105, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0908, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5154, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1677, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1642, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1264, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1647, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0842, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2845, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1548, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1671, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1315, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2300, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1350, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0761, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1280, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0779, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0130, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4649122807017544, 'recall': 0.4736371760500447, 'f1': 0.4692341744134572, 'number': 1119}, 'P': {'precision': 0.5142975549108993, 'recall': 0.6639914392723382, 'f1': 0.5796356842596917, 'number': 1869}, 'overall_precision': 0.4984520123839009, 'overall_recall': 0.5927041499330656, 'overall_f1': 0.541507414768384, 'overall_accuracy': 0.8199069440311629}
			------------EPOCH 11---------------
Loss:  tensor(0.2488, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3132, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1123, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1837, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0711, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4946, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0265, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0784, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0709, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0657, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3176, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0973, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1848, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1659, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0771, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0464, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0410, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0709, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0589, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0070, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.45542635658914726, 'recall': 0.42001787310098304, 'f1': 0.4370060437006043, 'number': 1119}, 'P': {'precision': 0.5856179775280899, 'recall': 0.6971642589620117, 'f1': 0.6365412799218367, 'number': 1869}, 'overall_precision': 0.5443659809640774, 'overall_recall': 0.5933734939759037, 'overall_f1': 0.5678142514011211, 'overall_accuracy': 0.8114805096437363}
			------------EPOCH 12---------------
Loss:  tensor(0.1045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0528, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0562, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1918, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0606, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5692, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0196, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0513, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0423, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0576, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0474, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0301, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2524, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0680, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1383, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1182, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1947, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0800, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0403, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0315, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0584, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0663, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0039, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5079197465681098, 'recall': 0.4298480786416443, 'f1': 0.46563407550822844, 'number': 1119}, 'P': {'precision': 0.6164502164502165, 'recall': 0.7619047619047619, 'f1': 0.6815027518545107, 'number': 1869}, 'overall_precision': 0.5848940743015044, 'overall_recall': 0.6375502008032129, 'overall_f1': 0.610088070456365, 'overall_accuracy': 0.823396542862553}
			------------EPOCH 13---------------
Loss:  tensor(0.0981, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3251, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0536, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0464, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1455, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0505, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4761, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0198, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0857, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0754, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0497, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2194, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0795, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0974, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0599, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0581, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0588, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0470, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0768, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0484, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0084, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.42590027700831024, 'recall': 0.5495978552278821, 'f1': 0.47990635973468593, 'number': 1119}, 'P': {'precision': 0.6678681771369722, 'recall': 0.6939539860888175, 'f1': 0.6806612437680398, 'number': 1869}, 'overall_precision': 0.5646780862374483, 'overall_recall': 0.6398929049531459, 'overall_f1': 0.5999372450580482, 'overall_accuracy': 0.817878107501285}
			------------EPOCH 14---------------
Loss:  tensor(0.1718, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2584, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0720, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0578, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0954, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0332, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3603, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0468, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0479, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0512, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0361, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0371, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2971, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0674, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1400, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0779, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0941, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0408, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0276, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0155, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0319, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0035, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4300760678759509, 'recall': 0.6568364611260054, 'f1': 0.5198019801980197, 'number': 1119}, 'P': {'precision': 0.6662797446314568, 'recall': 0.6142322097378277, 'f1': 0.6391982182628062, 'number': 1869}, 'overall_precision': 0.5486596736596736, 'overall_recall': 0.6301874163319946, 'overall_f1': 0.5866043613707166, 'overall_accuracy': 0.8117510211810534}
			------------EPOCH 15---------------
Loss:  tensor(0.1181, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1988, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0701, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0529, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0383, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4198, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0380, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0484, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0392, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0523, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0838, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0765, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0483, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0659, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0227, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0242, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0123, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0371, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0446, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0056, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.38720731010850945, 'recall': 0.6058981233243967, 'f1': 0.47247386759581883, 'number': 1119}, 'P': {'precision': 0.6082783443311338, 'recall': 0.5425361155698234, 'f1': 0.5735294117647057, 'number': 1869}, 'overall_precision': 0.4950263311878291, 'overall_recall': 0.5662650602409639, 'overall_f1': 0.5282547611614111, 'overall_accuracy': 0.7915167581897368}
			------------EPOCH 16---------------
Loss:  tensor(0.1460, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0850, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1255, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0384, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3785, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0286, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0185, 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(0.0254, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0312, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0997, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0417, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0324, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0622, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0317, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0390, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4497167138810198, 'recall': 0.5674709562109026, 'f1': 0.5017779533781114, 'number': 1119}, 'P': {'precision': 0.6230654018971543, 'recall': 0.6677367576243981, 'f1': 0.6446280991735538, 'number': 1869}, 'overall_precision': 0.5513909224011713, 'overall_recall': 0.6301874163319946, 'overall_f1': 0.5881617991566452, 'overall_accuracy': 0.8242892309356994}
			------------EPOCH 17---------------
Loss:  tensor(0.0577, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1345, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0334, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0412, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1157, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0306, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3396, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0328, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0558, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0276, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1621, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0152, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0585, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0526, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0640, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0421, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0571, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0270, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0655, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0478, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0178, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5886990801576872, 'recall': 0.40035746201966044, 'f1': 0.4765957446808511, 'number': 1119}, 'P': {'precision': 0.6131634819532908, 'recall': 0.7726056714820759, 'f1': 0.6837121212121212, 'number': 1869}, 'overall_precision': 0.6071887034659821, 'overall_recall': 0.6331994645247657, 'overall_f1': 0.6199213630406292, 'overall_accuracy': 0.8151459409743825}
			------------EPOCH 18---------------
Loss:  tensor(0.0671, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1432, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0255, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0687, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0302, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0159, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0194, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0532, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0341, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2426, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0452, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1203, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0538, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1293, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0169, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0130, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0218, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0691, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3978671041837572, 'recall': 0.4334226988382484, 'f1': 0.41488451668092385, 'number': 1119}, 'P': {'precision': 0.6098776619845945, 'recall': 0.7201712145532371, 'f1': 0.66045142296369, 'number': 1869}, 'overall_precision': 0.5344424985405721, 'overall_recall': 0.6127844712182061, 'overall_f1': 0.5709385718740256, 'overall_accuracy': 0.8167419590445533}
			------------EPOCH 19---------------
Loss:  tensor(0.0401, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1234, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0319, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0856, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0489, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2929, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0137, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0183, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0184, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0142, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1460, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0453, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0336, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0434, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0123, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0315, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.48522550544323484, 'recall': 0.5576407506702413, 'f1': 0.5189189189189188, 'number': 1119}, 'P': {'precision': 0.6511392405063291, 'recall': 0.6880684858212948, 'f1': 0.6690946930280957, 'number': 1869}, 'overall_precision': 0.5857099049371358, 'overall_recall': 0.6392235609103079, 'overall_f1': 0.6112978076492238, 'overall_accuracy': 0.8240051938215165}
			------------EPOCH 20---------------
Loss:  tensor(0.0329, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0932, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0163, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0455, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2679, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0227, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1270, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0387, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0197, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0295, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0085, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.47761194029850745, 'recall': 0.514745308310992, 'f1': 0.495483870967742, 'number': 1119}, 'P': {'precision': 0.6188026192703461, 'recall': 0.7078651685393258, 'f1': 0.6603443973047167, 'number': 1869}, 'overall_precision': 0.5678827751196173, 'overall_recall': 0.6355421686746988, 'overall_f1': 0.5998104864181933, 'overall_accuracy': 0.8275488949603701}
			------------EPOCH 21---------------
Loss:  tensor(0.0262, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0568, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0269, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2300, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, 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(0.0037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0252, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0135, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0176, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0171, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5201535508637236, 'recall': 0.484361036639857, 'f1': 0.5016196205460435, 'number': 1119}, 'P': {'precision': 0.6318181818181818, 'recall': 0.7437132156233279, 'f1': 0.6832145490292455, 'number': 1869}, 'overall_precision': 0.5959284392350401, 'overall_recall': 0.6465863453815262, 'overall_f1': 0.6202247191011236, 'overall_accuracy': 0.8278599832282847}
			------------EPOCH 22---------------
Loss:  tensor(0.0219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0434, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1955, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0099, 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(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0928, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0144, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0051, 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(0.0045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0188, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.47466216216216217, 'recall': 0.5022341376228776, 'f1': 0.4880590534085975, 'number': 1119}, 'P': {'precision': 0.6366650876361913, 'recall': 0.7191011235955056, 'f1': 0.6753768844221105, 'number': 1869}, 'overall_precision': 0.5784522003034901, 'overall_recall': 0.6378848728246319, 'overall_f1': 0.6067165366862964, 'overall_accuracy': 0.826534476695431}
			------------EPOCH 23---------------
Loss:  tensor(0.0192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0352, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0076, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0170, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1774, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0031, 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(0.0820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0213, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0127, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4918772563176895, 'recall': 0.4870420017873101, 'f1': 0.48944768747193534, 'number': 1119}, 'P': {'precision': 0.6282578875171467, 'recall': 0.7351524879614767, 'f1': 0.6775147928994083, 'number': 1869}, 'overall_precision': 0.5823975720789074, 'overall_recall': 0.642235609103079, 'overall_f1': 0.6108546872513131, 'overall_accuracy': 0.8279546622663456}
			------------EPOCH 24---------------
Loss:  tensor(0.0160, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0325, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0186, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1702, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0085, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, 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(0.0177, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0107, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4855390008764242, 'recall': 0.49508489722966936, 'f1': 0.4902654867256638, 'number': 1119}, 'P': {'precision': 0.6261072261072261, 'recall': 0.7185660781166399, 'f1': 0.669157947184853, 'number': 1869}, 'overall_precision': 0.5772976262933658, 'overall_recall': 0.6348728246318608, 'overall_f1': 0.6047178833280205, 'overall_accuracy': 0.8266697324640896}
			------------EPOCH 25---------------
Loss:  tensor(0.0137, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0344, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1725, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0918, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0169, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0185, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0035, 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(0.0021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4928825622775801, 'recall': 0.49508489722966936, 'f1': 0.4939812750780206, 'number': 1119}, 'P': {'precision': 0.6211573236889693, 'recall': 0.7351524879614767, 'f1': 0.673364371477579, 'number': 1869}, 'overall_precision': 0.5779376498800959, 'overall_recall': 0.6452476572958501, 'overall_f1': 0.6097406704617331, 'overall_accuracy': 0.8282522249573945}
			------------EPOCH 26---------------
Loss:  tensor(0.0112, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0475, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0214, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1783, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0090, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, 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(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.47215686274509805, 'recall': 0.5379803395889187, 'f1': 0.5029239766081871, 'number': 1119}, 'P': {'precision': 0.6512207274539114, 'recall': 0.6993044408774746, 'f1': 0.6744066047471621, 'number': 1869}, 'overall_precision': 0.5816575258988421, 'overall_recall': 0.6388888888888888, 'overall_f1': 0.6089314194577352, 'overall_accuracy': 0.8266426813103579}
			------------EPOCH 27---------------
Loss:  tensor(0.0083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0391, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0115, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0268, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2544, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0020, 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(0.0047, 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(0.1021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0139, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0236, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0270, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5264187866927593, 'recall': 0.48078641644325293, 'f1': 0.5025688930406352, 'number': 1119}, 'P': {'precision': 0.6107916850950906, 'recall': 0.7388978063135366, 'f1': 0.6687651331719129, 'number': 1869}, 'overall_precision': 0.5845263478525738, 'overall_recall': 0.642235609103079, 'overall_f1': 0.6120236007016425, 'overall_accuracy': 0.8264803743879676}
			------------EPOCH 28---------------
Loss:  tensor(0.0070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0457, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0183, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1506, 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(0.0069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0933, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0270, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0131, 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(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0183, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.46481178396072015, 'recall': 0.5075960679177838, 'f1': 0.48526270824434004, 'number': 1119}, 'P': {'precision': 0.6291390728476821, 'recall': 0.7116104868913857, 'f1': 0.6678383128295255, 'number': 1869}, 'overall_precision': 0.5689448441247003, 'overall_recall': 0.6352074966532798, 'overall_f1': 0.6002530044275775, 'overall_accuracy': 0.8280899180350042}
			------------EPOCH 29---------------
Loss:  tensor(0.0062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0251, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0135, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0020, 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(0.0628, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0107, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0017, 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(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5074349442379182, 'recall': 0.4879356568364611, 'f1': 0.49749430523918, 'number': 1119}, 'P': {'precision': 0.6200913242009133, 'recall': 0.7265917602996255, 'f1': 0.6691303276669132, 'number': 1869}, 'overall_precision': 0.5829761175750153, 'overall_recall': 0.6372155287817939, 'overall_f1': 0.6088903102014711, 'overall_accuracy': 0.826548002272297}
			------------EPOCH 30---------------
Loss:  tensor(0.0052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0212, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0065, 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(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0527, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0058, 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(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4632535094962841, 'recall': 0.5013404825737265, 'f1': 0.4815450643776824, 'number': 1119}, 'P': {'precision': 0.6181139122315593, 'recall': 0.7084002140181915, 'f1': 0.6601844926452256, 'number': 1869}, 'overall_precision': 0.562183119594393, 'overall_recall': 0.6308567603748326, 'overall_f1': 0.5945434474057719, 'overall_accuracy': 0.82810344361187}


		-------------RUN 4-----------
			------------EPOCH 1---------------
Loss:  tensor(3.2771, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6137, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3956, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5163, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2316, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0208, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1807, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8588, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9343, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2266, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7673, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6579, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7725, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8818, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5419, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2933, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5788, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5928, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5572, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7166, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1823, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5713, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.25, 'recall': 0.0009033423667570009, 'f1': 0.0018001800180018, 'number': 1107}, 'P': {'precision': 0.407293796864349, 'recall': 0.6296101159114857, 'f1': 0.49461920529801323, 'number': 1898}, 'overall_precision': 0.40707964601769914, 'overall_recall': 0.39800332778702163, 'overall_f1': 0.40249032475180885, 'overall_accuracy': 0.7019830644394003}
			------------EPOCH 2---------------
Loss:  tensor(2.0907, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5248, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2804, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0867, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0882, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5461, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9838, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1326, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9171, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9505, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2395, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6921, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8326, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6410, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8572, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8382, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7458, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4198, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9548, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6210, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4274, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.1, 'recall': 0.02077687443541102, 'f1': 0.03440538519072551, 'number': 1107}, 'P': {'precision': 0.4162453656892484, 'recall': 0.6506849315068494, 'f1': 0.5077081192189107, 'number': 1898}, 'overall_precision': 0.39349390053174854, 'overall_recall': 0.4186356073211314, 'overall_f1': 0.4056755885198323, 'overall_accuracy': 0.7183749473312221}
			------------EPOCH 3---------------
Loss:  tensor(1.8081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6661, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1141, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8745, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5656, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6443, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5584, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4907, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6181, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0462, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3905, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5679, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5259, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5929, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4710, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5321, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3332, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5467, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0339, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4352, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3556, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.1811764705882353, 'recall': 0.06955736224028906, 'f1': 0.10052219321148825, 'number': 1107}, 'P': {'precision': 0.45511265164644715, 'recall': 0.6917808219178082, 'f1': 0.549027806815806, 'number': 1898}, 'overall_precision': 0.4199395770392749, 'overall_recall': 0.46256239600665555, 'overall_f1': 0.440221694378464, 'overall_accuracy': 0.7354056515297731}
			------------EPOCH 4---------------
Loss:  tensor(1.4871, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6518, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3804, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3483, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0698, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3223, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2373, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0803, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4510, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2547, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8477, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0695, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2076, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4508, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4814, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2898, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2973, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9346, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3313, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3280, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.24523663183773817, 'recall': 0.3604336043360434, 'f1': 0.2918800292611558, 'number': 1107}, 'P': {'precision': 0.43345543345543347, 'recall': 0.5611169652265543, 'f1': 0.48909299655568317, 'number': 1898}, 'overall_precision': 0.3584720861900098, 'overall_recall': 0.4871880199667221, 'overall_f1': 0.4130342784595853, 'overall_accuracy': 0.7782474549087302}
			------------EPOCH 5---------------
Loss:  tensor(1.3880, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5145, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1287, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7690, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0784, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0914, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0691, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0520, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6781, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9160, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9717, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3823, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2701, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0629, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2293, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7616, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1203, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2951, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.260430994956442, 'recall': 0.5130984643179766, 'f1': 0.3454987834549878, 'number': 1107}, 'P': {'precision': 0.41120650700406686, 'recall': 0.4794520547945205, 'f1': 0.442714667963999, 'number': 1898}, 'overall_precision': 0.33636777423759673, 'overall_recall': 0.491846921797005, 'overall_f1': 0.39951344776321124, 'overall_accuracy': 0.79045301945007}
			------------EPOCH 6---------------
Loss:  tensor(1.2352, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4188, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9406, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6461, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8619, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8203, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8775, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6825, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8959, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0345, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2409, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7575, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5401, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6964, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7314, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0596, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9521, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7821, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1841, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7201, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0415, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2539, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4112, 'recall': 0.4643179765130985, 'f1': 0.43614764531183714, 'number': 1107}, 'P': {'precision': 0.553556827473426, 'recall': 0.7133825079030558, 'f1': 0.6233885819521179, 'number': 1898}, 'overall_precision': 0.5054112554112554, 'overall_recall': 0.621630615640599, 'overall_f1': 0.5575287270556633, 'overall_accuracy': 0.8147146371630897}
			------------EPOCH 7---------------
Loss:  tensor(1.1801, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2936, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7329, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6355, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6239, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6917, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4671, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7933, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0451, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9668, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3701, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5509, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2128, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7276, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4960, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4696, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5761, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1940, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.45491143317230276, 'recall': 0.5103884372177055, 'f1': 0.4810557684120903, 'number': 1107}, 'P': {'precision': 0.6466892218463025, 'recall': 0.7049525816649105, 'f1': 0.6745651625913789, 'number': 1898}, 'overall_precision': 0.574750830564784, 'overall_recall': 0.6332778702163061, 'overall_f1': 0.6025965801139962, 'overall_accuracy': 0.8102428880159841}
			------------EPOCH 8---------------
Loss:  tensor(1.0170, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2214, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6594, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4621, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5370, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4653, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4774, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3551, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4473, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3941, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0337, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6699, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4916, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3293, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4363, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4381, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1524, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8154, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7228, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3713, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1227, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3141, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3948, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6597, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1722, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.38206979542719616, 'recall': 0.5736224028906955, 'f1': 0.45864933188876855, 'number': 1107}, 'P': {'precision': 0.6076810889645115, 'recall': 0.6585879873551106, 'f1': 0.6321112515802781, 'number': 1898}, 'overall_precision': 0.5068566819037376, 'overall_recall': 0.627287853577371, 'overall_f1': 0.5606781677572874, 'overall_accuracy': 0.8183980536338059}
			------------EPOCH 9---------------
Loss:  tensor(0.9438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3993, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2525, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3470, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3233, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3865, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3325, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2955, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1523, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2696, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2478, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1118, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5829, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6251, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2422, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1267, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2910, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6257, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1480, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4817204301075269, 'recall': 0.4046973803071364, 'f1': 0.43986254295532645, 'number': 1107}, 'P': {'precision': 0.5817882159044088, 'recall': 0.743940990516333, 'f1': 0.6529479768786127, 'number': 1898}, 'overall_precision': 0.5540661304736372, 'overall_recall': 0.6189683860232945, 'overall_f1': 0.5847217856020119, 'overall_accuracy': 0.8179359275821293}
			------------EPOCH 10---------------
Loss:  tensor(0.7510, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1373, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2584, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1834, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1779, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1954, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1888, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2107, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1861, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0115, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5290, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2560, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1567, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1764, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1570, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5776, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0872, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1218, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1289, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3998082454458293, 'recall': 0.37669376693766937, 'f1': 0.387906976744186, 'number': 1107}, 'P': {'precision': 0.5680885972108285, 'recall': 0.7297154899894626, 'f1': 0.6388376383763836, 'number': 1898}, 'overall_precision': 0.5176673369721344, 'overall_recall': 0.5996672212978369, 'overall_f1': 0.5556583410422449, 'overall_accuracy': 0.8133554428934527}
			------------EPOCH 11---------------
Loss:  tensor(0.6145, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1194, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1793, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1297, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2566, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2317, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2275, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1867, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1115, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1190, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1591, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0737, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1787, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1315, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0930, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0974, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0835, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0843, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2952, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4965, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1289, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4135135135135135, 'recall': 0.4146341463414634, 'f1': 0.4140730717185386, 'number': 1107}, 'P': {'precision': 0.5890298812934917, 'recall': 0.7581664910432033, 'f1': 0.6629808799815711, 'number': 1898}, 'overall_precision': 0.5341964537010977, 'overall_recall': 0.6316139767054908, 'overall_f1': 0.5788350106739859, 'overall_accuracy': 0.817188370733829}
			------------EPOCH 12---------------
Loss:  tensor(0.5775, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0896, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1956, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0657, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0695, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0982, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1267, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0953, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0539, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0647, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3598, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0805, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0380, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1154, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0772, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0593, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3353, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5384, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1712, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0715, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2708, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4406, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1059, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.47042766151046406, 'recall': 0.4670280036133695, 'f1': 0.4687216681776972, 'number': 1107}, 'P': {'precision': 0.6255545696539485, 'recall': 0.7428872497365648, 'f1': 0.6791907514450866, 'number': 1898}, 'overall_precision': 0.574709215627796, 'overall_recall': 0.6412645590682197, 'overall_f1': 0.606165460836741, 'overall_accuracy': 0.8296657741290963}
			------------EPOCH 13---------------
Loss:  tensor(0.4910, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0773, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0637, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0305, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0564, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0785, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1796, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0733, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0933, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3812, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0898, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0275, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0940, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0882, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0527, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3869, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4958, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0829, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0942, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0784, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2607, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4222, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1315, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.46785109983079526, 'recall': 0.4995483288166215, 'f1': 0.4831804281345565, 'number': 1107}, 'P': {'precision': 0.595945945945946, 'recall': 0.6970495258166491, 'f1': 0.6425449247207382, 'number': 1898}, 'overall_precision': 0.551440329218107, 'overall_recall': 0.6242928452579035, 'overall_f1': 0.5856094896207273, 'overall_accuracy': 0.814904924360839}
			------------EPOCH 14---------------
Loss:  tensor(0.5562, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0819, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1632, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0546, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1242, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0769, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0963, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0887, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0689, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1593, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4645, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0871, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0411, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0489, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3120, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4287, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0949, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0960, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1932, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3087, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4952, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1278, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.42820347714101736, 'recall': 0.6007226738934056, 'f1': 0.5, 'number': 1107}, 'P': {'precision': 0.6299582463465553, 'recall': 0.6359325605900948, 'f1': 0.6329313057157839, 'number': 1898}, 'overall_precision': 0.5396367829345633, 'overall_recall': 0.6229617304492513, 'overall_f1': 0.5783132530120483, 'overall_accuracy': 0.8169844915933835}
			------------EPOCH 15---------------
Loss:  tensor(0.4556, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0722, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0790, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0711, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0458, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0688, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0565, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0716, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0298, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0638, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0139, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7267, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1218, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0483, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0350, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3922, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4132, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0225, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0724, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0549, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1656, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3489483747609943, 'recall': 0.6594399277326106, 'f1': 0.4563926226945921, 'number': 1107}, 'P': {'precision': 0.6483739837398373, 'recall': 0.5042149631190727, 'f1': 0.5672791938352103, 'number': 1898}, 'overall_precision': 0.4728139013452915, 'overall_recall': 0.5613976705490848, 'overall_f1': 0.5133120340788073, 'overall_accuracy': 0.7677001073763473}
			------------EPOCH 16---------------
Loss:  tensor(0.6843, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3629, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2809, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0722, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1709, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0680, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0380, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0673, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3868, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2228, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1803, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2837, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0974, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5272, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0672, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0550, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2812, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4735, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1301, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.371960682876358, 'recall': 0.6495031616982837, 'f1': 0.47302631578947374, 'number': 1107}, 'P': {'precision': 0.6324567030147531, 'recall': 0.5194942044257113, 'f1': 0.5704367949088806, 'number': 1898}, 'overall_precision': 0.48825887743413515, 'overall_recall': 0.56738768718802, 'overall_f1': 0.524857626596891, 'overall_accuracy': 0.7840376224973835}
			------------EPOCH 17---------------
Loss:  tensor(0.5675, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0873, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2613, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2542, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4294, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3584, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2084, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1646, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1681, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0537, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0809, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0661, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0859, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0790, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5396, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5763, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3266, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1412, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2910, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5433, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1397, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.45564821834723274, 'recall': 0.5429087624209575, 'f1': 0.4954657873042045, 'number': 1107}, 'P': {'precision': 0.6202709014479215, 'recall': 0.6996838777660696, 'f1': 0.6575885120079228, 'number': 1898}, 'overall_precision': 0.557514450867052, 'overall_recall': 0.6419301164725457, 'overall_f1': 0.5967517401392111, 'overall_accuracy': 0.8291356883639378}
			------------EPOCH 18---------------
Loss:  tensor(0.4416, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0503, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0449, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0356, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0426, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0418, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0757, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1396, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1818, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0258, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5492, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1780, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0395, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0665, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0672, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3462, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0539, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0299, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2184, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3988, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1045, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.44188034188034186, 'recall': 0.4670280036133695, 'f1': 0.4541062801932367, 'number': 1107}, 'P': {'precision': 0.6276881720430108, 'recall': 0.738145416227608, 'f1': 0.678450363196126, 'number': 1898}, 'overall_precision': 0.5637860082304527, 'overall_recall': 0.6382695507487521, 'overall_f1': 0.5987201498361168, 'overall_accuracy': 0.8228018430674297}
			------------EPOCH 19---------------
Loss:  tensor(0.4022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0453, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0481, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0197, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0415, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0450, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0787, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0790, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0372, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2467, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0230, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0240, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0303, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2444, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3387, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0152, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0498, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1967, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3549, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0869, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.43769968051118213, 'recall': 0.4950316169828365, 'f1': 0.4646036456125477, 'number': 1107}, 'P': {'precision': 0.6123646209386282, 'recall': 0.7149631190727082, 'f1': 0.6596985901798736, 'number': 1898}, 'overall_precision': 0.5493079584775087, 'overall_recall': 0.6339434276206323, 'overall_f1': 0.588598794994593, 'overall_accuracy': 0.8260095415437728}
			------------EPOCH 20---------------
Loss:  tensor(0.3576, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0306, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0176, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0207, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0650, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0742, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0098, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0177, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2639, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0175, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0259, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2455, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3382, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0084, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0452, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1860, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3307, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0833, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.459546925566343, 'recall': 0.5130984643179766, 'f1': 0.48484848484848486, 'number': 1107}, 'P': {'precision': 0.6365348399246704, 'recall': 0.7123287671232876, 'f1': 0.6723023371456986, 'number': 1898}, 'overall_precision': 0.5714285714285714, 'overall_recall': 0.6389351081530782, 'overall_f1': 0.6032992930086409, 'overall_accuracy': 0.8272735922145352}
			------------EPOCH 21---------------
Loss:  tensor(0.3316, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0246, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0098, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0141, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0181, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0640, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0321, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0120, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0107, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3196, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0346, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1744, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2947, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0723, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4509493670886076, 'recall': 0.5149051490514905, 'f1': 0.48080978490088566, 'number': 1107}, 'P': {'precision': 0.633147113594041, 'recall': 0.7165437302423604, 'f1': 0.6722689075630252, 'number': 1898}, 'overall_precision': 0.5656506447831184, 'overall_recall': 0.6422628951747088, 'overall_f1': 0.601527193392551, 'overall_accuracy': 0.8295162627594362}
			------------EPOCH 22---------------
Loss:  tensor(0.3089, 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(0.0056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0143, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0517, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0284, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2320, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3373, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0475, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1705, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2853, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0725, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.48016528925619834, 'recall': 0.5248419150858176, 'f1': 0.5015105740181268, 'number': 1107}, 'P': {'precision': 0.6479471448796602, 'recall': 0.7233930453108536, 'f1': 0.6835947224296739, 'number': 1898}, 'overall_precision': 0.5869630519675578, 'overall_recall': 0.6502495840266223, 'overall_f1': 0.6169876855067887, 'overall_accuracy': 0.8311880717110897}
			------------EPOCH 23---------------
Loss:  tensor(0.2936, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0174, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0144, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0391, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0335, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0044, 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(0.2167, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0076, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0213, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2194, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2980, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0391, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1930, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2881, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0701, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4752724224643755, 'recall': 0.5121951219512195, 'f1': 0.4930434782608696, 'number': 1107}, 'P': {'precision': 0.627110908261068, 'recall': 0.7239199157007377, 'f1': 0.6720469552457814, 'number': 1898}, 'overall_precision': 0.5735815602836879, 'overall_recall': 0.6459234608985025, 'overall_f1': 0.6076068242291438, 'overall_accuracy': 0.8281978443178883}
			------------EPOCH 24---------------
Loss:  tensor(0.2719, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0195, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0427, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1765, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0017, 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(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1778, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0295, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1659, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2773, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0682, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4393700787401575, 'recall': 0.5040650406504065, 'f1': 0.4694993689524611, 'number': 1107}, 'P': {'precision': 0.639584317430326, 'recall': 0.7133825079030558, 'f1': 0.6744707347447073, 'number': 1898}, 'overall_precision': 0.5645113669914379, 'overall_recall': 0.6362728785357737, 'overall_f1': 0.5982478097622028, 'overall_accuracy': 0.8297473257852744}
			------------EPOCH 25---------------
Loss:  tensor(0.2741, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0271, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0396, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0032, 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(0.1671, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0318, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1785, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2713, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0357, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2465, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0671, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.47732893652102226, 'recall': 0.5230352303523035, 'f1': 0.4991379310344828, 'number': 1107}, 'P': {'precision': 0.6336861481822366, 'recall': 0.7255005268703899, 'f1': 0.6764922623434045, 'number': 1898}, 'overall_precision': 0.5776727702303603, 'overall_recall': 0.6509151414309484, 'overall_f1': 0.6121107807854795, 'overall_accuracy': 0.8312832153099643}
			------------EPOCH 26---------------
Loss:  tensor(0.2492, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0133, 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(0.0068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0310, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0329, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1887, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0074, 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(0.0197, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1745, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2788, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1526, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2619, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0657, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4648105181747873, 'recall': 0.5429087624209575, 'f1': 0.5008333333333332, 'number': 1107}, 'P': {'precision': 0.6536394723986322, 'recall': 0.7049525816649105, 'f1': 0.6783269961977186, 'number': 1898}, 'overall_precision': 0.5805389221556886, 'overall_recall': 0.6452579034941763, 'overall_f1': 0.6111899133175729, 'overall_accuracy': 0.8319899963301755}
			------------EPOCH 27---------------
Loss:  tensor(0.2559, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0114, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0271, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0264, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1346, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0182, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1344, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2145, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1336, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2283, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0608, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.49151103565365023, 'recall': 0.5230352303523035, 'f1': 0.5067833698030634, 'number': 1107}, 'P': {'precision': 0.6447803495512517, 'recall': 0.7191780821917808, 'f1': 0.6799501867995018, 'number': 1898}, 'overall_precision': 0.5899848254931714, 'overall_recall': 0.6469217970049916, 'overall_f1': 0.6171428571428571, 'overall_accuracy': 0.830793905372895}
			------------EPOCH 28---------------
Loss:  tensor(0.2281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0224, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0260, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1324, 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(0.0024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0195, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1464, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1945, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0521, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.46434108527131784, 'recall': 0.5411020776874436, 'f1': 0.49979140592407173, 'number': 1107}, 'P': {'precision': 0.6438159156279961, 'recall': 0.7075869336143309, 'f1': 0.6741967871485944, 'number': 1898}, 'overall_precision': 0.5752369668246445, 'overall_recall': 0.6462562396006656, 'overall_f1': 0.6086820247610092, 'overall_accuracy': 0.8300463485245946}
			------------EPOCH 29---------------
Loss:  tensor(0.2054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0158, 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(0.0058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0042, 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(0.0023, 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(0.1249, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0230, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1210, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1856, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0972, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1810, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0418, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.47815333882934874, 'recall': 0.5239385727190605, 'f1': 0.5, 'number': 1107}, 'P': {'precision': 0.6388758782201405, 'recall': 0.7186512118018967, 'f1': 0.6764195388048599, 'number': 1898}, 'overall_precision': 0.5806451612903226, 'overall_recall': 0.6469217970049916, 'overall_f1': 0.611994333385802, 'overall_accuracy': 0.8306308020605385}
			------------EPOCH 30---------------
Loss:  tensor(0.1934, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0132, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0196, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0316, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0020, 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(0.0874, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0184, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1298, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0190, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0798, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1612, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0369, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.47528830313014825, 'recall': 0.5212285456187895, 'f1': 0.4971994829814735, 'number': 1107}, 'P': {'precision': 0.6463356973995272, 'recall': 0.720231822971549, 'f1': 0.681285821081485, 'number': 1898}, 'overall_precision': 0.5839591468909583, 'overall_recall': 0.6469217970049916, 'overall_f1': 0.6138301231449322, 'overall_accuracy': 0.8304133309773966}


		-------------RUN 5-----------
			------------EPOCH 1---------------
Loss:  tensor(0.4456, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2380, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2972, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3328, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8507, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7302, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2947, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6195, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1593, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6150, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1527, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3273, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4927, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2586, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1528, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4876, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0958, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4760, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1191, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0273, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4102, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3140096618357488, 'recall': 0.11850501367365543, 'f1': 0.17207147584381205, 'number': 1097}, 'P': {'precision': 0.47831978319783197, 'recall': 0.5966197183098592, 'f1': 0.5309601403860618, 'number': 1775}, 'overall_precision': 0.4524353120243531, 'overall_recall': 0.4139972144846797, 'overall_f1': 0.43236363636363634, 'overall_accuracy': 0.7179537217730694}
			------------EPOCH 2---------------
Loss:  tensor(0.2642, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1434, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3631, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3202, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8171, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8956, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5556, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6646, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7591, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6150, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5970, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7931, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5531, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6666, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7521, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5647, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5717, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3690, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5478, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0321, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5631, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4889, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0387, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.34386617100371747, 'recall': 0.33728350045578853, 'f1': 0.34054302807179015, 'number': 1097}, 'P': {'precision': 0.40584415584415584, 'recall': 0.49295774647887325, 'f1': 0.4451793436784533, 'number': 1775}, 'overall_precision': 0.38521039603960394, 'overall_recall': 0.4334958217270195, 'overall_f1': 0.4079292267365662, 'overall_accuracy': 0.7430443267354335}
			------------EPOCH 3---------------
Loss:  tensor(0.2094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1173, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0668, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0199, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5142, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3449, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5962, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2376, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2741, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3996, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1904, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2796, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1368, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3838, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2430, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2804, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3164, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2890, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8543, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2446, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1927, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2411, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8409, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4215686274509804, 'recall': 0.47037374658158615, 'f1': 0.44463593278759156, 'number': 1097}, 'P': {'precision': 0.42208642514185946, 'recall': 0.5447887323943662, 'f1': 0.4756517461878997, 'number': 1775}, 'overall_precision': 0.42190611664295874, 'overall_recall': 0.5163649025069638, 'overall_f1': 0.46438077344606227, 'overall_accuracy': 0.7752996933370505}
			------------EPOCH 4---------------
Loss:  tensor(0.1416, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0836, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8237, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8271, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1806, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2795, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0475, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0787, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4362, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9660, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1273, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9578, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9777, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0985, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9111, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0768, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2324, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0619, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6420, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9614, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9242, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9483, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6646, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.42123056118999325, 'recall': 0.5679124886052871, 'f1': 0.483695652173913, 'number': 1097}, 'P': {'precision': 0.5242621310655328, 'recall': 0.5904225352112676, 'f1': 0.5553789083200849, 'number': 1775}, 'overall_precision': 0.480448533640023, 'overall_recall': 0.5818245125348189, 'overall_f1': 0.5262992125984252, 'overall_accuracy': 0.7982157792026763}
			------------EPOCH 5---------------
Loss:  tensor(0.0754, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0650, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6264, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7417, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0306, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1364, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8650, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9175, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7395, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9686, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7422, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7114, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6436, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7843, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7299, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9241, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1866, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8846, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7507, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6686, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7180, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4738, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4905802562170309, 'recall': 0.593436645396536, 'f1': 0.5371287128712872, 'number': 1097}, 'P': {'precision': 0.54345703125, 'recall': 0.6270422535211267, 'f1': 0.5822652367250851, 'number': 1775}, 'overall_precision': 0.5226666666666666, 'overall_recall': 0.6142061281337048, 'overall_f1': 0.564751080518649, 'overall_accuracy': 0.8158628380262057}
			------------EPOCH 6---------------
Loss:  tensor(0.0517, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0455, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4649, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5938, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7731, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9826, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5667, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7147, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6130, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5458, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5433, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4337, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5573, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5310, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7127, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6759, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3262, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5711, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4787, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2961, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5211822660098522, 'recall': 0.4822242479489517, 'f1': 0.5009469696969696, 'number': 1097}, 'P': {'precision': 0.5196285352469396, 'recall': 0.6935211267605634, 'f1': 0.5941119691119692, 'number': 1775}, 'overall_precision': 0.5200945626477541, 'overall_recall': 0.6128133704735376, 'overall_f1': 0.5626598465473145, 'overall_accuracy': 0.8201282408698076}
			------------EPOCH 7---------------
Loss:  tensor(0.0345, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0327, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5308, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5994, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3810, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5639, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2670, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5555, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8910, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4789, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3447, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3762, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4266, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6327, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7553, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1305, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6381, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2396, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3637, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3271, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3565, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.37615571776155715, 'recall': 0.7046490428441203, 'f1': 0.49048223350253806, 'number': 1097}, 'P': {'precision': 0.6027088036117382, 'recall': 0.4512676056338028, 'f1': 0.5161082474226804, 'number': 1775}, 'overall_precision': 0.4651300236406619, 'overall_recall': 0.5480501392757661, 'overall_f1': 0.5031969309462915, 'overall_accuracy': 0.7451212712573181}
			------------EPOCH 8---------------
Loss:  tensor(0.0607, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0242, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3525, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6825, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7875, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9286, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2934, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4526, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1804, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3339, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5586, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3552, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4212, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2274, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2746, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5435, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1257, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2510, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1508, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4324984728161271, 'recall': 0.6453965360072926, 'f1': 0.5179224579370886, 'number': 1097}, 'P': {'precision': 0.5833333333333334, 'recall': 0.579718309859155, 'f1': 0.5815202034473015, 'number': 1775}, 'overall_precision': 0.5107321376065863, 'overall_recall': 0.6048050139275766, 'overall_f1': 0.5538020086083215, 'overall_accuracy': 0.7915110119877334}
			------------EPOCH 9---------------
Loss:  tensor(0.0168, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0147, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4373, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5797, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7662, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2816, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4143, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1722, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2803, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2881, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2489, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2552, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2377, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2754, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2214, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0907, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4701, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2665, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2986, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3689, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2047, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.45815295815295815, 'recall': 0.5788514129443938, 'f1': 0.5114780507450664, 'number': 1097}, 'P': {'precision': 0.6154672395273899, 'recall': 0.6456338028169014, 'f1': 0.6301897167995602, 'number': 1775}, 'overall_precision': 0.5483374384236454, 'overall_recall': 0.620125348189415, 'overall_f1': 0.5820261437908496, 'overall_accuracy': 0.7929606913855589}
			------------EPOCH 10---------------
Loss:  tensor(0.0233, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0175, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2781, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4959, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6545, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9902, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3193, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5384, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1174, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2358, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4983, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2629, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2542, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3517, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4336, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3715, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4135, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6908, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6168, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1806, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3704, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2345, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2574, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1556, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4954591321897074, 'recall': 0.44758432087511396, 'f1': 0.4703065134099617, 'number': 1097}, 'P': {'precision': 0.5695807314897413, 'recall': 0.7194366197183099, 'f1': 0.63579785909883, 'number': 1775}, 'overall_precision': 0.5468605010825859, 'overall_recall': 0.6155988857938719, 'overall_f1': 0.5791973791973792, 'overall_accuracy': 0.8030945079453582}
			------------EPOCH 11---------------
Loss:  tensor(0.0179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0269, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2818, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4417, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6939, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0764, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2952, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5422, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1704, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3448, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6342, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3866, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3996, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2737, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1974, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1432, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5537, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4376, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1747, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2486, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3181, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1855, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1701, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5565669700910273, 'recall': 0.390154968094804, 'f1': 0.4587352625937835, 'number': 1097}, 'P': {'precision': 0.5483460559796438, 'recall': 0.7284507042253521, 'f1': 0.6256956206145656, 'number': 1775}, 'overall_precision': 0.5503677646306364, 'overall_recall': 0.599233983286908, 'overall_f1': 0.5737622937156193, 'overall_accuracy': 0.8074017284638975}
			------------EPOCH 12---------------
Loss:  tensor(0.0286, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0223, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2923, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4646, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5647, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8374, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2367, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3764, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1342, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1716, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4417, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2457, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1248, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1544, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1272, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1266, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1425, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5491, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1275, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4551, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1385, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1408, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1324, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0781, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4492656875834446, 'recall': 0.6134913400182316, 'f1': 0.5186897880539499, 'number': 1097}, 'P': {'precision': 0.6027173913043479, 'recall': 0.6247887323943662, 'f1': 0.6135546334716461, 'number': 1775}, 'overall_precision': 0.5338526063511084, 'overall_recall': 0.6204735376044568, 'overall_f1': 0.5739130434782609, 'overall_accuracy': 0.8117228882074157}
			------------EPOCH 13---------------
Loss:  tensor(0.0060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2974, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3593, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0713, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0695, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0710, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2317, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0839, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0901, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0781, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0711, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3456, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0632, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2984, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0542, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0720, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0667, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0382, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0294, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5225594749794914, 'recall': 0.5806745670009116, 'f1': 0.5500863557858376, 'number': 1097}, 'P': {'precision': 0.6339144215530903, 'recall': 0.676056338028169, 'f1': 0.6543075245365322, 'number': 1775}, 'overall_precision': 0.5902956298200515, 'overall_recall': 0.6396239554317549, 'overall_f1': 0.6139705882352942, 'overall_accuracy': 0.8204209645943685}
			------------EPOCH 14---------------
Loss:  tensor(0.0027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0970, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2673, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5240, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0511, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1521, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0609, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0424, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3396, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1295, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0415, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0507, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0544, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0433, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0478, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3199, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0550, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2734, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0384, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0408, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0324, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0228, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0192, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5228426395939086, 'recall': 0.5633546034639927, 'f1': 0.5423431329530496, 'number': 1097}, 'P': {'precision': 0.6244366549824737, 'recall': 0.7025352112676057, 'f1': 0.66118769883351, 'number': 1775}, 'overall_precision': 0.5866624724756213, 'overall_recall': 0.6493732590529248, 'overall_f1': 0.6164270368534127, 'overall_accuracy': 0.8210621689434068}
			------------EPOCH 15---------------
Loss:  tensor(0.0021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0830, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2397, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2816, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0356, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1307, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0481, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0350, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3288, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1084, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0317, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0444, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0338, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0303, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0412, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3310, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0589, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2531, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0296, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0305, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0199, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0165, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5063795853269537, 'recall': 0.5788514129443938, 'f1': 0.540195661420672, 'number': 1097}, 'P': {'precision': 0.6180878552971576, 'recall': 0.6738028169014084, 'f1': 0.644743935309973, 'number': 1775}, 'overall_precision': 0.5741611790529947, 'overall_recall': 0.6375348189415042, 'overall_f1': 0.6041907276027059, 'overall_accuracy': 0.8214803456927795}
			------------EPOCH 16---------------
Loss:  tensor(0.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0778, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2203, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2539, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4444, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0333, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1139, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0437, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0268, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2972, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0929, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0242, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0264, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0237, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0385, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2595, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0482, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2102, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0244, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0208, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0128, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0116, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5223628691983122, 'recall': 0.5642661804922516, 'f1': 0.5425065731814198, 'number': 1097}, 'P': {'precision': 0.6496350364963503, 'recall': 0.7019718309859155, 'f1': 0.6747901435147577, 'number': 1775}, 'overall_precision': 0.6010312600708991, 'overall_recall': 0.6493732590529248, 'overall_f1': 0.6242677824267782, 'overall_accuracy': 0.8215221633677168}
			------------EPOCH 17---------------
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0660, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2252, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0187, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0880, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0342, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2794, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0690, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0180, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0345, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2383, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0421, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1727, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0196, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0157, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0093, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5298570227081582, 'recall': 0.5742935278030994, 'f1': 0.5511811023622046, 'number': 1097}, 'P': {'precision': 0.635336414997432, 'recall': 0.6969014084507043, 'f1': 0.6646963997850618, 'number': 1775}, 'overall_precision': 0.595344387755102, 'overall_recall': 0.6500696378830083, 'overall_f1': 0.6215046604527297, 'overall_accuracy': 0.822470030666295}
			------------EPOCH 18---------------
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0422, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1748, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1937, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3539, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0148, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0663, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0175, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2527, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0580, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0150, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0324, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2401, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1504, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0171, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0123, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0075, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5301102629346904, 'recall': 0.5697356426618049, 'f1': 0.5492091388400703, 'number': 1097}, 'P': {'precision': 0.6335372069317023, 'recall': 0.7002816901408451, 'f1': 0.6652394969226653, 'number': 1775}, 'overall_precision': 0.59471505889844, 'overall_recall': 0.6504178272980501, 'overall_f1': 0.621320472309995, 'overall_accuracy': 0.8229021466406468}
			------------EPOCH 19---------------
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0367, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1650, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1854, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3363, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0546, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0238, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2312, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0518, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0085, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0090, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0353, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1269, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0135, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0057, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.528830313014827, 'recall': 0.585232452142206, 'f1': 0.5556036347901341, 'number': 1097}, 'P': {'precision': 0.6499733617474693, 'recall': 0.6873239436619718, 'f1': 0.6681270536692223, 'number': 1775}, 'overall_precision': 0.6023940472339049, 'overall_recall': 0.6483286908077994, 'overall_f1': 0.6245178601375146, 'overall_accuracy': 0.820643992194034}
			------------EPOCH 20---------------
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0509, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1541, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1752, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3353, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0443, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0211, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2243, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0475, 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(0.0089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0264, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0436, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1473, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0102, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0048, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5088141025641025, 'recall': 0.5788514129443938, 'f1': 0.5415778251599147, 'number': 1097}, 'P': {'precision': 0.6473072861668426, 'recall': 0.6907042253521126, 'f1': 0.6683019896429544, 'number': 1775}, 'overall_precision': 0.5922978994271165, 'overall_recall': 0.6479805013927576, 'overall_f1': 0.6188892583970735, 'overall_accuracy': 0.8207276275439086}
			------------EPOCH 21---------------
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0568, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1613, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1819, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3311, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0688, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0300, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2346, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0475, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1838, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0318, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0040, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5360915492957746, 'recall': 0.5551504102096627, 'f1': 0.5454545454545453, 'number': 1097}, 'P': {'precision': 0.627977698935631, 'recall': 0.6980281690140845, 'f1': 0.6611526147278548, 'number': 1775}, 'overall_precision': 0.5944033451270505, 'overall_recall': 0.6434540389972145, 'overall_f1': 0.6179568634007692, 'overall_accuracy': 0.8219682185670477}
			------------EPOCH 22---------------
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0554, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1556, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1728, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0307, 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(0.2046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0436, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0170, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2156, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0410, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1455, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0077, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0037, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5248107653490328, 'recall': 0.568824065633546, 'f1': 0.5459317585301837, 'number': 1097}, 'P': {'precision': 0.6367804381049414, 'recall': 0.704225352112676, 'f1': 0.6688068485821295, 'number': 1775}, 'overall_precision': 0.5945431472081218, 'overall_recall': 0.6525069637883009, 'overall_f1': 0.6221779548472776, 'overall_accuracy': 0.8227627543908559}
			------------EPOCH 23---------------
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0283, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1195, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1398, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2695, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0350, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0193, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0107, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0466, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1578, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0260, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0934, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0031, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5190397350993378, 'recall': 0.5715587967183227, 'f1': 0.5440347071583513, 'number': 1097}, 'P': {'precision': 0.6544973544973545, 'recall': 0.6969014084507043, 'f1': 0.6750341064120055, 'number': 1775}, 'overall_precision': 0.6016785022595222, 'overall_recall': 0.649025069637883, 'overall_f1': 0.6244556113902848, 'overall_accuracy': 0.8236688040144968}
			------------EPOCH 24---------------
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0267, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2575, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0115, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1708, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0284, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0084, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1439, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0255, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0811, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0026, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5334507042253521, 'recall': 0.5524156791248861, 'f1': 0.542767577250336, 'number': 1097}, 'P': {'precision': 0.6254399195575666, 'recall': 0.7008450704225352, 'f1': 0.660998937300744, 'number': 1775}, 'overall_precision': 0.592, 'overall_recall': 0.6441504178272981, 'overall_f1': 0.6169751542437886, 'overall_accuracy': 0.8226651798160022}
			------------EPOCH 25---------------
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0266, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0834, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1139, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0159, 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(0.1621, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0243, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1336, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0258, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0708, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0058, 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(0.0024, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5126324368378158, 'recall': 0.5733819507748404, 'f1': 0.5413080895008605, 'number': 1097}, 'P': {'precision': 0.6303219106957425, 'recall': 0.683943661971831, 'f1': 0.6560389084031343, 'number': 1775}, 'overall_precision': 0.584522676815731, 'overall_recall': 0.6417130919220055, 'overall_f1': 0.6117842323651452, 'overall_accuracy': 0.8202815723445777}
			------------EPOCH 26---------------
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0309, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0695, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0160, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0111, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1535, 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(0.0025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1150, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0602, 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(0.0020, 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>)
				 {'C': {'precision': 0.5186721991701245, 'recall': 0.5697356426618049, 'f1': 0.5430060816681146, 'number': 1097}, 'P': {'precision': 0.6430446194225722, 'recall': 0.6901408450704225, 'f1': 0.6657608695652174, 'number': 1775}, 'overall_precision': 0.594855305466238, 'overall_recall': 0.6441504178272981, 'overall_f1': 0.618522233366767, 'overall_accuracy': 0.8221215500418176}
			------------EPOCH 27---------------
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0260, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0792, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0952, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1870, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0129, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1379, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0264, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1077, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0666, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0019, 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(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5176663927691043, 'recall': 0.5742935278030994, 'f1': 0.5445116681071738, 'number': 1097}, 'P': {'precision': 0.6321243523316062, 'recall': 0.6873239436619718, 'f1': 0.6585695006747638, 'number': 1775}, 'overall_precision': 0.5878614553543057, 'overall_recall': 0.6441504178272981, 'overall_f1': 0.6147200531649776, 'overall_accuracy': 0.8222470030666295}
			------------EPOCH 28---------------
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0289, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0671, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0815, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1680, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1417, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0242, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0915, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0174, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0505, 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(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5333926981300089, 'recall': 0.5460346399270738, 'f1': 0.5396396396396397, 'number': 1097}, 'P': {'precision': 0.6199407699901284, 'recall': 0.7076056338028169, 'f1': 0.6608787161273351, 'number': 1775}, 'overall_precision': 0.5890758971101937, 'overall_recall': 0.645891364902507, 'overall_f1': 0.6161767148314234, 'overall_accuracy': 0.8242124337886814}
			------------EPOCH 29---------------
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0412, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0757, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0114, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1348, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0224, 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(0.0039, 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(0.1843, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1140, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0024, 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(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5182186234817814, 'recall': 0.5834092980856882, 'f1': 0.5488850771869639, 'number': 1097}, 'P': {'precision': 0.6381201044386423, 'recall': 0.6884507042253522, 'f1': 0.662330623306233, 'number': 1775}, 'overall_precision': 0.5911111111111111, 'overall_recall': 0.6483286908077994, 'overall_f1': 0.6183992029226171, 'overall_accuracy': 0.822010036241985}
			------------EPOCH 30---------------
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0481, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0562, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0942, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1646, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1625, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0398, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1499, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0193, 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(0.0056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, 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>)
				 {'C': {'precision': 0.5181219110378913, 'recall': 0.5733819507748404, 'f1': 0.5443530938987452, 'number': 1097}, 'P': {'precision': 0.6446886446886447, 'recall': 0.6940845070422536, 'f1': 0.6684753119913186, 'number': 1775}, 'overall_precision': 0.59552, 'overall_recall': 0.6479805013927576, 'overall_f1': 0.6206436551609138, 'overall_accuracy': 0.8225257875662113}
