Tokenizer: ../arg_m/arg_mining/smlm_pretrained_iter5_0/tokenizer Model: ../arg_m/arg_mining/smlm_pretrained_iter5_0/model
	Train size: 80 Test size: 20


		-------------RUN 1-----------
			------------EPOCH 1---------------
Loss:  tensor(3.5608, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6495, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1874, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0489, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7490, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6303, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7704, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7499, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8695, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1962, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4342, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4725, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7769, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3165, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9726, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2149, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2774, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1392, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1362, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4478, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4210, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9847, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9527, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8249, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8473, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9679, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7481, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6872, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.24193548387096775, 'recall': 0.07042253521126761, 'f1': 0.10909090909090909, 'number': 426}, 'P': {'precision': 0.3864353312302839, 'recall': 0.625, 'f1': 0.4775828460038986, 'number': 784}, 'overall_precision': 0.3735632183908046, 'overall_recall': 0.4297520661157025, 'overall_f1': 0.39969254419677175, 'overall_accuracy': 0.7257944349065466}
			------------EPOCH 2---------------
Loss:  tensor(1.7537, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7532, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2914, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5714, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6308, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4926, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3978, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3581, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0672, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0977, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5746, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1361, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9839, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3272, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1300, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5148, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3287, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2899, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6441, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5292, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5991, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2907, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4249, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2665, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1929, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3223, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9157, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4191, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3923, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3413, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1748, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2340, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2739, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2553, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2926, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2538, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2874493927125506, 'recall': 0.3333333333333333, 'f1': 0.30869565217391304, 'number': 426}, 'P': {'precision': 0.5334040296924708, 'recall': 0.6415816326530612, 'f1': 0.5825130283729009, 'number': 784}, 'overall_precision': 0.4488517745302714, 'overall_recall': 0.5330578512396694, 'overall_f1': 0.4873441632036267, 'overall_accuracy': 0.7651841534125878}
			------------EPOCH 3---------------
Loss:  tensor(1.1528, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2600, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2145, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1513, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0808, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0839, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2871, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0643, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2539, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7679, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2208, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0839, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7891, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2986, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1811, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7710, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3732, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1682, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2715, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5500, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2238, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4721, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0807, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9837, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1368, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9220, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6943, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3378, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0978, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1886, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0153, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9462, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9454, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9937, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1361, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9630, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0387, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.27157894736842103, 'recall': 0.3028169014084507, 'f1': 0.2863485016648169, 'number': 426}, 'P': {'precision': 0.5426587301587301, 'recall': 0.6977040816326531, 'f1': 0.6104910714285714, 'number': 784}, 'overall_precision': 0.45583277140930545, 'overall_recall': 0.5586776859504132, 'overall_f1': 0.5020423319717787, 'overall_accuracy': 0.7886365844967769}
			------------EPOCH 4---------------
Loss:  tensor(0.9841, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0641, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9985, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8196, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8298, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8837, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8569, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5259, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6935, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9441, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0634, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6330, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9499, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3130, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2475, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3249, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9758, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2698, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5374, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7552, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1238, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6939, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5825, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2718, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5128, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7584, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8892, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6847, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6761, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6601, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6778, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8428, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6331, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7987, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.38524590163934425, 'recall': 0.5516431924882629, 'f1': 0.45366795366795365, 'number': 426}, 'P': {'precision': 0.6113033448673587, 'recall': 0.6760204081632653, 'f1': 0.6420351302241066, 'number': 784}, 'overall_precision': 0.5179417738659445, 'overall_recall': 0.6322314049586777, 'overall_f1': 0.5694082620022329, 'overall_accuracy': 0.8005895500631661}
			------------EPOCH 5---------------
Loss:  tensor(0.6360, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7282, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1444, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6197, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5835, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1668, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5184, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6200, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5199, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7190, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0415, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3943, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6818, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5485, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5184, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1793, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6880, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1856, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5231, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0853, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4572, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3358, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3859, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5262, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6187, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4343, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4307, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4247, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3897, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5911, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3940594059405941, 'recall': 0.4671361502347418, 'f1': 0.4274973147153598, 'number': 426}, 'P': {'precision': 0.6065934065934065, 'recall': 0.7040816326530612, 'f1': 0.6517119244391972, 'number': 784}, 'overall_precision': 0.5307420494699646, 'overall_recall': 0.6206611570247934, 'overall_f1': 0.5721904761904762, 'overall_accuracy': 0.7986459784263549}
			------------EPOCH 6---------------
Loss:  tensor(0.4283, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4288, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0893, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4264, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3544, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3518, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1372, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2167, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4336, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4969, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0287, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2417, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4660, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2312, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1241, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4944, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1628, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2531, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2949, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0508, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2588, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2284, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1828, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2597, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4414, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2507, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2373, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2369, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4974, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3391, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5567, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.47115384615384615, 'recall': 0.460093896713615, 'f1': 0.4655581947743468, 'number': 426}, 'P': {'precision': 0.5495495495495496, 'recall': 0.7002551020408163, 'f1': 0.6158160403813797, 'number': 784}, 'overall_precision': 0.5265017667844523, 'overall_recall': 0.6157024793388429, 'overall_f1': 0.5676190476190476, 'overall_accuracy': 0.8078131579799812}
			------------EPOCH 7---------------
Loss:  tensor(0.2269, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2420, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0721, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2718, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3979, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1930, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2243, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1320, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2427, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5448, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3579, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1534, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0926, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2426, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0964, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0922, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5493, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1527, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1788, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1953, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0293, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2445, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1783, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2185, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1327, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2277, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2808, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1752, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2102, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1948, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2393, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4738, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2520, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4094, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3517730496453901, 'recall': 0.5821596244131455, 'f1': 0.43854995579133504, 'number': 426}, 'P': {'precision': 0.6452442159383034, 'recall': 0.6403061224489796, 'f1': 0.6427656850192062, 'number': 784}, 'overall_precision': 0.5057316250842886, 'overall_recall': 0.6198347107438017, 'overall_f1': 0.5569996286669142, 'overall_accuracy': 0.7908716918791099}
			------------EPOCH 8---------------
Loss:  tensor(0.2421, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2305, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0553, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2701, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2357, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1874, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0803, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3333, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1185, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5830, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0170, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2892, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0475, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3714, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1202, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0878, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7370, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1429, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1760, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1772, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1836, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0136, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1331, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1641, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0927, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1584, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3335, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2391, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2506, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2274, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3968, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1957, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4642, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3514285714285714, 'recall': 0.5774647887323944, 'f1': 0.43694493783303734, 'number': 426}, 'P': {'precision': 0.6711229946524064, 'recall': 0.6403061224489796, 'f1': 0.6553524804177546, 'number': 784}, 'overall_precision': 0.5165745856353591, 'overall_recall': 0.6181818181818182, 'overall_f1': 0.5628291948833708, 'overall_accuracy': 0.7836480839622947}
			------------EPOCH 9---------------
Loss:  tensor(0.1659, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1904, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0252, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1624, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3402, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2927, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2451, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0896, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2828, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3085, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0076, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1317, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2325, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1481, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1814, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0348, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2000, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0620, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0501, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2929, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0875, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1644, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1461, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1651, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0968, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1314, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0979, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2105, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1402, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1497, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2615, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4713, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3751891074130106, 'recall': 0.5821596244131455, 'f1': 0.4563017479300828, 'number': 426}, 'P': {'precision': 0.6909814323607427, 'recall': 0.6645408163265306, 'f1': 0.6775032509752926, 'number': 784}, 'overall_precision': 0.5434628975265018, 'overall_recall': 0.6355371900826446, 'overall_f1': 0.5859047619047619, 'overall_accuracy': 0.7974474425836545}
			------------EPOCH 10---------------
Loss:  tensor(0.1618, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0849, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0139, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1218, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1598, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1318, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1422, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1259, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1877, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0786, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3156, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3946, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0650, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2937, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0975, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2203, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0334, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2239, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0564, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3430, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0634, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0795, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0612, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1201, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0076, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0535, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0685, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1565, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0643, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0690, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2257, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0939, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0794, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0887, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1181, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1666, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3398, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4407894736842105, 'recall': 0.47183098591549294, 'f1': 0.4557823129251701, 'number': 426}, 'P': {'precision': 0.6097826086956522, 'recall': 0.7155612244897959, 'f1': 0.6584507042253521, 'number': 784}, 'overall_precision': 0.5537790697674418, 'overall_recall': 0.6297520661157024, 'overall_f1': 0.5893271461716938, 'overall_accuracy': 0.8066470149978945}
			------------EPOCH 11---------------
Loss:  tensor(0.1648, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0420, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2906, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1340, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0616, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0924, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1382, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2321, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0690, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2501, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1356, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0853, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1412, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0705, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0287, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0635, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1703, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0588, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0547, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0358, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1482, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0596, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0425, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0386, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0589, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2448, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0904, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2702, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4375, 'recall': 0.460093896713615, 'f1': 0.448512585812357, 'number': 426}, 'P': {'precision': 0.6069868995633187, 'recall': 0.7091836734693877, 'f1': 0.6541176470588235, 'number': 784}, 'overall_precision': 0.5513196480938416, 'overall_recall': 0.6214876033057851, 'overall_f1': 0.5843045843045844, 'overall_accuracy': 0.8146804444300476}
			------------EPOCH 12---------------
Loss:  tensor(0.0947, 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>)
Loss:  tensor(0.0726, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0791, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0790, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0510, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0502, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1730, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0784, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1892, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0077, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0477, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0419, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0956, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0204, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1114, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0602, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0130, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1500, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0242, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0457, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0408, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0711, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0366, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1348, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0940, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0373, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0692, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0305, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0715, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2418, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0714, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2552, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.45558086560364464, 'recall': 0.4694835680751174, 'f1': 0.46242774566473993, 'number': 426}, 'P': {'precision': 0.6248648648648648, 'recall': 0.7372448979591837, 'f1': 0.6764189584552369, 'number': 784}, 'overall_precision': 0.5703812316715543, 'overall_recall': 0.6429752066115703, 'overall_f1': 0.6045066045066045, 'overall_accuracy': 0.822227980952998}
			------------EPOCH 13---------------
Loss:  tensor(0.0591, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0643, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0328, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0584, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0569, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0343, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0520, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0930, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0153, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1768, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1764, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0275, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0580, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0246, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0862, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0498, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0433, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0077, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0185, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0300, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0169, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0364, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0341, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0870, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0145, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0793, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0536, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0295, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0222, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0237, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1935, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0679, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2399, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4037735849056604, 'recall': 0.5023474178403756, 'f1': 0.44769874476987453, 'number': 426}, 'P': {'precision': 0.6487119437939111, 'recall': 0.7066326530612245, 'f1': 0.6764346764346764, 'number': 784}, 'overall_precision': 0.5549132947976878, 'overall_recall': 0.6347107438016529, 'overall_f1': 0.5921356977640708, 'overall_accuracy': 0.8057724077613294}
			------------EPOCH 14---------------
Loss:  tensor(0.0461, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0239, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0331, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0275, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0355, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0623, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1437, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1594, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0199, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0383, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0195, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0539, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0408, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0403, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0102, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1334, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0272, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0197, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0366, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0136, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0938, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0139, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0615, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0202, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0208, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0087, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0148, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1857, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0488, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2051, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.37320574162679426, 'recall': 0.5492957746478874, 'f1': 0.4444444444444444, 'number': 426}, 'P': {'precision': 0.674775928297055, 'recall': 0.6721938775510204, 'f1': 0.673482428115016, 'number': 784}, 'overall_precision': 0.5404829545454546, 'overall_recall': 0.6289256198347107, 'overall_f1': 0.5813598166539342, 'overall_accuracy': 0.8048006219429238}
			------------EPOCH 15---------------
Loss:  tensor(0.0442, 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(0.0398, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0202, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0177, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0313, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0184, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0543, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1711, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1777, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0164, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0601, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0190, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0351, 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(0.0374, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0876, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0187, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0171, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0181, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0107, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0890, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0522, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2811, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.36808846761453395, 'recall': 0.5469483568075117, 'f1': 0.44003777148253065, 'number': 426}, 'P': {'precision': 0.6427718040621266, 'recall': 0.6862244897959183, 'f1': 0.6637877853177051, 'number': 784}, 'overall_precision': 0.5244897959183673, 'overall_recall': 0.6371900826446281, 'overall_f1': 0.5753731343283582, 'overall_accuracy': 0.8047682290823103}
			------------EPOCH 16---------------
Loss:  tensor(0.0441, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0133, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0169, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0424, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0120, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0411, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2120, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2182, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0340, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0762, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0383, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0140, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0569, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0993, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0238, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0145, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0114, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0464, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2418, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0684, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2208, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4074074074074074, 'recall': 0.4647887323943662, 'f1': 0.43421052631578944, 'number': 426}, 'P': {'precision': 0.6280353200883002, 'recall': 0.7257653061224489, 'f1': 0.6733727810650888, 'number': 784}, 'overall_precision': 0.5510057471264368, 'overall_recall': 0.6338842975206611, 'overall_f1': 0.5895465026902383, 'overall_accuracy': 0.8163000874607237}
			------------EPOCH 17---------------
Loss:  tensor(0.0383, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0135, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0313, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0076, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0316, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0090, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0504, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1383, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1426, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0335, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0654, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0132, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0307, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0846, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0133, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0129, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0149, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0939, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0373, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0076, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1841, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0401, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2187, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3938356164383562, 'recall': 0.539906103286385, 'f1': 0.4554455445544555, 'number': 426}, 'P': {'precision': 0.6367924528301887, 'recall': 0.6887755102040817, 'f1': 0.6617647058823529, 'number': 784}, 'overall_precision': 0.5377094972067039, 'overall_recall': 0.6363636363636364, 'overall_f1': 0.582891748675246, 'overall_accuracy': 0.8054808720158078}
			------------EPOCH 18---------------
Loss:  tensor(0.0290, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0112, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0159, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0280, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0510, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1242, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1257, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0198, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0239, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0282, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0535, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0149, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0107, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0684, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0333, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0085, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1568, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0279, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1748, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.39718804920913886, 'recall': 0.5305164319248826, 'f1': 0.4542713567839196, 'number': 426}, 'P': {'precision': 0.6547619047619048, 'recall': 0.701530612244898, 'f1': 0.6773399014778325, 'number': 784}, 'overall_precision': 0.5507452093683464, 'overall_recall': 0.6413223140495867, 'overall_f1': 0.5925925925925924, 'overall_accuracy': 0.8069061578828026}
			------------EPOCH 19---------------
Loss:  tensor(0.0202, 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(0.0026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0326, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0942, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0854, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0194, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0200, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0311, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0267, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0524, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0117, 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(0.0068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0608, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0287, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1265, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0200, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1420, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.431640625, 'recall': 0.5187793427230047, 'f1': 0.47121535181236673, 'number': 426}, 'P': {'precision': 0.649365628604383, 'recall': 0.7181122448979592, 'f1': 0.6820109024833435, 'number': 784}, 'overall_precision': 0.5685279187817259, 'overall_recall': 0.6479338842975206, 'overall_f1': 0.6056392429509464, 'overall_accuracy': 0.8139030157753231}
			------------EPOCH 20---------------
Loss:  tensor(0.0185, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0199, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0345, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0853, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0831, 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(0.0104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0305, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0107, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0225, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0406, 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(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0571, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0236, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0023, 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(0.0051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0142, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1349, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.38605442176870747, 'recall': 0.5328638497652582, 'f1': 0.4477317554240632, 'number': 426}, 'P': {'precision': 0.6365740740740741, 'recall': 0.701530612244898, 'f1': 0.6674757281553398, 'number': 784}, 'overall_precision': 0.5351239669421488, 'overall_recall': 0.6421487603305785, 'overall_f1': 0.583771600300526, 'overall_accuracy': 0.8104369796896764}
			------------EPOCH 21---------------
Loss:  tensor(0.0163, 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(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0174, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0255, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0621, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0631, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0331, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0234, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0459, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0606, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0205, 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(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1137, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1459, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.40325497287522605, 'recall': 0.5234741784037559, 'f1': 0.45556690500510727, 'number': 426}, 'P': {'precision': 0.6468531468531469, 'recall': 0.7079081632653061, 'f1': 0.6760048721071864, 'number': 784}, 'overall_precision': 0.5513819985825655, 'overall_recall': 0.6429752066115703, 'overall_f1': 0.5936665394887447, 'overall_accuracy': 0.8086553723559328}
			------------EPOCH 22---------------
Loss:  tensor(0.0120, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0089, 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(0.0035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0170, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0225, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0582, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0593, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0384, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0544, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0023, 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(0.0056, 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(0.0026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0523, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, 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(0.0015, 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(0.0042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1197, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4329501915708812, 'recall': 0.5305164319248826, 'f1': 0.4767932489451477, 'number': 426}, 'P': {'precision': 0.6256983240223464, 'recall': 0.7142857142857143, 'f1': 0.6670637284097678, 'number': 784}, 'overall_precision': 0.5546930134086098, 'overall_recall': 0.6495867768595042, 'overall_f1': 0.5984012181195281, 'overall_accuracy': 0.814388908684526}
			------------EPOCH 23---------------
Loss:  tensor(0.0092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0025, 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(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0549, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0503, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0307, 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(0.0211, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0346, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0034, 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(0.0025, 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(0.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0160, 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(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1210, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1010, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4066543438077634, 'recall': 0.5164319248826291, 'f1': 0.4550155118924509, 'number': 426}, 'P': {'precision': 0.6266968325791855, 'recall': 0.7066326530612245, 'f1': 0.6642685851318945, 'number': 784}, 'overall_precision': 0.5431578947368421, 'overall_recall': 0.6396694214876033, 'overall_f1': 0.5874762808349145, 'overall_accuracy': 0.8120566227203524}
			------------EPOCH 24---------------
Loss:  tensor(0.0071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0181, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0609, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0496, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0299, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0314, 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(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, 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(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0151, 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(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0137, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0954, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3879310344827586, 'recall': 0.528169014084507, 'f1': 0.4473161033797216, 'number': 426}, 'P': {'precision': 0.6468489892984542, 'recall': 0.6938775510204082, 'f1': 0.6695384615384615, 'number': 784}, 'overall_precision': 0.5411681914144968, 'overall_recall': 0.6355371900826446, 'overall_f1': 0.5845686050931204, 'overall_accuracy': 0.8030838003304072}
			------------EPOCH 25---------------
Loss:  tensor(0.0059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0072, 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(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0147, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0213, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0540, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0457, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0116, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, 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(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0230, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0343, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0470, 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(0.0139, 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(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0149, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1023, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4329501915708812, 'recall': 0.5305164319248826, 'f1': 0.4767932489451477, 'number': 426}, 'P': {'precision': 0.625421822272216, 'recall': 0.7091836734693877, 'f1': 0.6646742378959953, 'number': 784}, 'overall_precision': 0.5542168674698795, 'overall_recall': 0.6462809917355372, 'overall_f1': 0.596718809614651, 'overall_accuracy': 0.8144860872663665}
			------------EPOCH 26---------------
Loss:  tensor(0.0053, 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(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0163, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0507, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0336, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0243, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0218, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0327, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0373, 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(0.0126, 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(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0821, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0087, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0758, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3896551724137931, 'recall': 0.5305164319248826, 'f1': 0.4493041749502982, 'number': 426}, 'P': {'precision': 0.6240090600226501, 'recall': 0.7028061224489796, 'f1': 0.6610677864427115, 'number': 784}, 'overall_precision': 0.5311004784688995, 'overall_recall': 0.6421487603305785, 'overall_f1': 0.5813692480359146, 'overall_accuracy': 0.8096919438955654}
			------------EPOCH 27---------------
Loss:  tensor(0.0051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0062, 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(0.0043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0204, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0394, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0242, 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(0.0179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0371, 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(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0390, 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(0.0111, 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(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0732, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0855, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.40142095914742454, 'recall': 0.5305164319248826, 'f1': 0.4570273003033367, 'number': 426}, 'P': {'precision': 0.6316397228637414, 'recall': 0.6977040816326531, 'f1': 0.663030303030303, 'number': 784}, 'overall_precision': 0.5409377186843947, 'overall_recall': 0.6388429752066116, 'overall_f1': 0.5858279651383099, 'overall_accuracy': 0.8105665511321305}
			------------EPOCH 28---------------
Loss:  tensor(0.0049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0085, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0143, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0446, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0255, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0236, 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(0.0177, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0394, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0376, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0102, 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(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0639, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0782, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3931034482758621, 'recall': 0.5352112676056338, 'f1': 0.4532803180914513, 'number': 426}, 'P': {'precision': 0.6421800947867299, 'recall': 0.6913265306122449, 'f1': 0.665847665847666, 'number': 784}, 'overall_precision': 0.5407303370786517, 'overall_recall': 0.6363636363636364, 'overall_f1': 0.5846621108580106, 'overall_accuracy': 0.8080723008648893}
			------------EPOCH 29---------------
Loss:  tensor(0.0048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0057, 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(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0143, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0115, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0650, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0459, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.6088e-05, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0048, 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(0.0235, 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(0.0204, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0280, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, 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(0.0001, 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(0.0354, 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(0.0099, 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(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0897, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0240, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0942, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4173076923076923, 'recall': 0.5093896713615024, 'f1': 0.45877378435517974, 'number': 426}, 'P': {'precision': 0.6108695652173913, 'recall': 0.7168367346938775, 'f1': 0.65962441314554, 'number': 784}, 'overall_precision': 0.5409722222222222, 'overall_recall': 0.643801652892562, 'overall_f1': 0.5879245283018868, 'overall_accuracy': 0.8177901590489456}
			------------EPOCH 30---------------
Loss:  tensor(0.0047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, 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(0.0124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0098, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0341, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0471, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0982e-05, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0246, 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(0.0163, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0244, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0001, 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(0.0376, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, 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(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0933, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0141, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0817, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4339622641509434, 'recall': 0.539906103286385, 'f1': 0.48117154811715485, 'number': 426}, 'P': {'precision': 0.6204217536071032, 'recall': 0.7130102040816326, 'f1': 0.6635014836795251, 'number': 784}, 'overall_precision': 0.5513626834381551, 'overall_recall': 0.6520661157024793, 'overall_f1': 0.5975009466111322, 'overall_accuracy': 0.8155226588059992}


		-------------RUN 2-----------
			------------EPOCH 1---------------
Loss:  tensor(3.6051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5801, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0450, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1501, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8836, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8955, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9474, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3481, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7357, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6434, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4547, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6993, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5671, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4759, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5493, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1857, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4186, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9917, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2645, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1840, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2517, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2916, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3107, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8540, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4699, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1465, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0355, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6503, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3529, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9641, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9267, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7896, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6726, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8224, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2971698113207547, 'recall': 0.11954459203036052, 'f1': 0.1705006765899865, 'number': 527}, 'P': {'precision': 0.32319134550371875, 'recall': 0.5241228070175439, 'f1': 0.3998327059807612, 'number': 912}, 'overall_precision': 0.3199290360733294, 'overall_recall': 0.37595552466990967, 'overall_f1': 0.34568690095846644, 'overall_accuracy': 0.7360486048604861}
			------------EPOCH 2---------------
Loss:  tensor(1.5995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7582, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1741, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5400, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2300, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5803, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6461, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7498, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5614, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0259, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4367, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9204, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5875, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3978, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1156, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2569, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5988, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4385, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0866, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1715, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2077, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4671, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3988, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1674, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8701, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4499, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2430, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1791, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3139, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7962, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2074, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.39946380697050937, 'recall': 0.2827324478178368, 'f1': 0.3311111111111111, 'number': 527}, 'P': {'precision': 0.41721854304635764, 'recall': 0.6217105263157895, 'f1': 0.4993394980184941, 'number': 912}, 'overall_precision': 0.4133949191685912, 'overall_recall': 0.4975677553856845, 'overall_f1': 0.4515925575528224, 'overall_accuracy': 0.7961296129612961}
			------------EPOCH 3---------------
Loss:  tensor(1.0992, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4906, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8735, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0849, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8323, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1315, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1350, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0605, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6393, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3144, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6562, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1658, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1331, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9470, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9876, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0746, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1739, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1321, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1123, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9301, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1695, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9112, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2121, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1421, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0744, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0321, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7765, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1684, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9427, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1168, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9504, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5688, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8598, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5054945054945055, 'recall': 0.4364326375711575, 'f1': 0.4684317718940937, 'number': 527}, 'P': {'precision': 0.4972332015810277, 'recall': 0.6896929824561403, 'f1': 0.5778594395957741, 'number': 912}, 'overall_precision': 0.4994186046511628, 'overall_recall': 0.5969423210562891, 'overall_f1': 0.5438429882874327, 'overall_accuracy': 0.8323582358235824}
			------------EPOCH 4---------------
Loss:  tensor(0.7522, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7811, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7540, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0243, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9296, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2757, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4814, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7684, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6168, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6815, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0534, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4380, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2309, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0292, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5555, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0646, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0965, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6884, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2404, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0496, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8837, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6864, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2307, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5926, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4867, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0227, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7220, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7919, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4187, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6500, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4394736842105263, 'recall': 0.6337760910815939, 'f1': 0.519036519036519, 'number': 527}, 'P': {'precision': 0.5691489361702128, 'recall': 0.5866228070175439, 'f1': 0.5777537796976242, 'number': 912}, 'overall_precision': 0.5111764705882353, 'overall_recall': 0.6038915913829048, 'overall_f1': 0.5536795157693534, 'overall_accuracy': 0.8280546804680468}
			------------EPOCH 5---------------
Loss:  tensor(0.6386, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2557, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5590, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5614, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4571, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7296, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9143, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7853, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3852, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1951, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3436, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5258, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7231, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5482, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0240, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0983, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7924, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0197, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3758, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0663, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5566, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5846, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1456, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7396, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6509, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5274, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1806, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4613, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3717, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0450, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6376, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6454, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7729, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6144, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2783, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4531, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5244956772334294, 'recall': 0.34535104364326374, 'f1': 0.41647597254004576, 'number': 527}, 'P': {'precision': 0.5049354593773728, 'recall': 0.7291666666666666, 'f1': 0.5966801256168686, 'number': 912}, 'overall_precision': 0.5090144230769231, 'overall_recall': 0.5886031966643502, 'overall_f1': 0.5459233000322268, 'overall_accuracy': 0.8173098559855986}
			------------EPOCH 6---------------
Loss:  tensor(0.5996, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2337, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6602, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5765, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5844, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7551, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4511, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6448, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5994, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2257, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1793, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2716, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5284, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6289, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4999, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0198, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0990, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1839, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7725, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0272, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2994, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0320, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3107, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0734, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1280, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6933, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6878, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1763, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6518, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7984, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4467, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3749, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3680, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.42117376294591485, 'recall': 0.6944971537001897, 'f1': 0.5243553008595989, 'number': 527}, 'P': {'precision': 0.6368563685636857, 'recall': 0.5153508771929824, 'f1': 0.5696969696969697, 'number': 912}, 'overall_precision': 0.5202240199128811, 'overall_recall': 0.580958999305073, 'overall_f1': 0.5489166119500984, 'overall_accuracy': 0.7937668766876688}
			------------EPOCH 7---------------
Loss:  tensor(0.5936, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2791, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5257, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2499, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6916, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2928, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4443, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5797, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1746, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1913, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3210, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4792, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5141, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3584, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3863, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0184, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0823, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1414, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5858, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0283, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2139, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0279, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1199, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4884, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1974, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1287, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7293, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2476, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5633, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2549, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1911, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5, 'recall': 0.5939278937381404, 'f1': 0.5429314830875976, 'number': 527}, 'P': {'precision': 0.618, 'recall': 0.6776315789473685, 'f1': 0.6464435146443515, 'number': 912}, 'overall_precision': 0.5725707257072571, 'overall_recall': 0.6469770674079222, 'overall_f1': 0.6075040783034258, 'overall_accuracy': 0.8400371287128713}
			------------EPOCH 8---------------
Loss:  tensor(0.2310, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3181, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1489, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1596, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5190, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1603, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4572, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1294, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1647, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1423, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2804, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1868, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1680, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0147, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0436, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0590, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4236, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1141, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0102, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0130, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0859, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3171, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3392, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1885, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0596, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0955, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0894, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5702, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1137, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4458, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1529, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1850, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1183, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5141509433962265, 'recall': 0.6204933586337761, 'f1': 0.5623387790197765, 'number': 527}, 'P': {'precision': 0.6584845250800427, 'recall': 0.6765350877192983, 'f1': 0.6673877771768524, 'number': 912}, 'overall_precision': 0.600127145581691, 'overall_recall': 0.6560111188325226, 'overall_f1': 0.6268260292164675, 'overall_accuracy': 0.8380681818181818}
			------------EPOCH 9---------------
Loss:  tensor(0.1219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0744, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2662, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0931, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1529, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3783, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2572, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3876, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0419, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0886, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0873, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1941, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0076, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0137, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0342, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2907, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0536, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0076, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0574, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0638, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2152, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2601, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1414, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0676, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0510, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4848, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0556, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1121, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1260, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0788, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4873015873015873, 'recall': 0.5825426944971537, 'f1': 0.5306828003457217, 'number': 527}, 'P': {'precision': 0.6418067226890757, 'recall': 0.6699561403508771, 'f1': 0.655579399141631, 'number': 912}, 'overall_precision': 0.5802781289506953, 'overall_recall': 0.6379430159833217, 'overall_f1': 0.6077457795431976, 'overall_accuracy': 0.8378431593159316}
			------------EPOCH 10---------------
Loss:  tensor(0.0546, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0415, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2157, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0492, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1183, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3274, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0909, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3857, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0343, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1421, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0643, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0761, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1239, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0515, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0750, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0242, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2669, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0381, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0373, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0356, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0514, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1797, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2155, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0373, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4505, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0383, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0745, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0775, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0979, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0688, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4903339191564148, 'recall': 0.5294117647058824, 'f1': 0.5091240875912408, 'number': 527}, 'P': {'precision': 0.6739367502726281, 'recall': 0.6776315789473685, 'f1': 0.675779114270093, 'number': 912}, 'overall_precision': 0.6036339165545087, 'overall_recall': 0.6233495482974287, 'overall_f1': 0.6133333333333332, 'overall_accuracy': 0.8297704770477048}
			------------EPOCH 11---------------
Loss:  tensor(0.0384, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0423, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2000, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0270, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2986, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0590, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1709, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3143, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0208, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0393, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0547, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0719, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0355, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0621, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0247, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2553, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0628, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0266, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0463, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0701, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1951, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0875, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0087, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0348, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0274, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0386, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0590, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3385, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0514, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0493, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0297, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5209205020920502, 'recall': 0.47248576850094876, 'f1': 0.4955223880597015, 'number': 527}, 'P': {'precision': 0.6504065040650406, 'recall': 0.7017543859649122, 'f1': 0.6751054852320675, 'number': 912}, 'overall_precision': 0.6080711354309165, 'overall_recall': 0.6177901320361362, 'overall_f1': 0.6128921061702861, 'overall_accuracy': 0.8263107560756076}
			------------EPOCH 12---------------
Loss:  tensor(0.0490, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0498, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2497, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0268, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1137, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1977, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0112, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0262, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0441, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0598, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0553, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0164, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2880, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0909, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0866, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0430, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0646, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1550, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1922, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0648, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0107, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0314, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0218, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4197, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0334, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0514, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3491, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0601, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0789, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0536, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.546236559139785, 'recall': 0.4819734345351044, 'f1': 0.5120967741935484, 'number': 527}, 'P': {'precision': 0.6314312441534145, 'recall': 0.7401315789473685, 'f1': 0.6814740030287733, 'number': 912}, 'overall_precision': 0.605606258148631, 'overall_recall': 0.645587213342599, 'overall_f1': 0.6249579549276825, 'overall_accuracy': 0.8362117461746175}
			------------EPOCH 13---------------
Loss:  tensor(0.0298, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0319, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1632, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0214, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0676, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2425, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0740, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1207, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0203, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1166, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0329, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1307, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0116, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0440, 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>)
Loss:  tensor(0.0037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2105, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0461, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0669, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0287, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0517, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2000, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1960, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0514, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0249, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3616, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0193, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0369, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2834, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0461, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0821, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0504, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5486111111111112, 'recall': 0.4497153700189753, 'f1': 0.4942648592283629, 'number': 527}, 'P': {'precision': 0.6041055718475073, 'recall': 0.6776315789473685, 'f1': 0.6387596899224807, 'number': 912}, 'overall_precision': 0.5876288659793815, 'overall_recall': 0.5941626129256428, 'overall_f1': 0.5908776779543884, 'overall_accuracy': 0.8180974347434743}
			------------EPOCH 14---------------
Loss:  tensor(0.0748, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0387, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1609, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0363, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0739, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2751, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1424, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2463, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0815, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0307, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0662, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0436, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0324, 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(0.0086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0514, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0282, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0592, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0460, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1622, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0665, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0199, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0451, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0268, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3463, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0672, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0492, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2497, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0592, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0152, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5398550724637681, 'recall': 0.5654648956356736, 'f1': 0.5523632993512512, 'number': 527}, 'P': {'precision': 0.6352705410821643, 'recall': 0.6951754385964912, 'f1': 0.6638743455497381, 'number': 912}, 'overall_precision': 0.6012903225806452, 'overall_recall': 0.6476719944405838, 'overall_f1': 0.6236199397791904, 'overall_accuracy': 0.8385463546354636}
			------------EPOCH 15---------------
Loss:  tensor(0.0183, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0238, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0454, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2620, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0464, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1387, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2186, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0655, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0345, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0628, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0337, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0440, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1932, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0126, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0297, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0403, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1371, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1639, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0814, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0173, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3212, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0165, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2455, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0307, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1264, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0146, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.45726495726495725, 'recall': 0.6091081593927894, 'f1': 0.5223759153783564, 'number': 527}, 'P': {'precision': 0.6152241918665277, 'recall': 0.6469298245614035, 'f1': 0.6306787814003206, 'number': 912}, 'overall_precision': 0.5484647802528597, 'overall_recall': 0.6330785267546908, 'overall_f1': 0.587741935483871, 'overall_accuracy': 0.8312893789378938}
			------------EPOCH 16---------------
Loss:  tensor(0.0610, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2351, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0332, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0847, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0115, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0186, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0168, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0111, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0320, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0629, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1592, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0176, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0084, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0358, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0991, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1531, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0464, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0366, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3222, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0165, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0247, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2386, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0327, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0476, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0163, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.46763754045307443, 'recall': 0.5483870967741935, 'f1': 0.5048034934497816, 'number': 527}, 'P': {'precision': 0.5302325581395348, 'recall': 0.625, 'f1': 0.5737292400603925, 'number': 912}, 'overall_precision': 0.5073833431777909, 'overall_recall': 0.5969423210562891, 'overall_f1': 0.5485312899106002, 'overall_accuracy': 0.833230198019802}
			------------EPOCH 17---------------
Loss:  tensor(0.0236, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0264, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1369, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0428, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2200, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0367, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1272, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2361, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0663, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0202, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0210, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0142, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0271, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1545, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0090, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0173, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0432, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0860, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0401, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0169, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0111, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3327, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0114, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0173, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0240, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0339, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0092, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.51985559566787, 'recall': 0.5464895635673624, 'f1': 0.5328399629972248, 'number': 527}, 'P': {'precision': 0.583941605839416, 'recall': 0.7017543859649122, 'f1': 0.6374501992031871, 'number': 912}, 'overall_precision': 0.5624242424242424, 'overall_recall': 0.6448922863099374, 'overall_f1': 0.6008416963418581, 'overall_accuracy': 0.8400933843384338}
			------------EPOCH 18---------------
Loss:  tensor(0.0064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0136, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1484, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0596, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0359, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1534, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2956, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0115, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0774, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0461, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0342, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0153, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, 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(0.1118, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0112, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0301, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0581, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1111, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0434, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0168, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2567, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0312, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2194, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0412, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1213, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0161, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5292153589315526, 'recall': 0.6015180265654649, 'f1': 0.5630550621669627, 'number': 527}, 'P': {'precision': 0.6698513800424628, 'recall': 0.6918859649122807, 'f1': 0.680690399137001, 'number': 912}, 'overall_precision': 0.6151849448410124, 'overall_recall': 0.6587908269631688, 'overall_f1': 0.6362416107382551, 'overall_accuracy': 0.8402340234023402}
			------------EPOCH 19---------------
Loss:  tensor(0.0047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0975, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0678, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1512, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0257, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1613, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0476, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0139, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0123, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0262, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1242, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0116, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0339, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0660, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0778, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0479, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0159, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1603, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0152, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0315, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0073, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5561224489795918, 'recall': 0.6204933586337761, 'f1': 0.5865470852017937, 'number': 527}, 'P': {'precision': 0.6887921653971708, 'recall': 0.694078947368421, 'f1': 0.6914254505734572, 'number': 912}, 'overall_precision': 0.6370272063702721, 'overall_recall': 0.6671299513551077, 'overall_f1': 0.6517311608961303, 'overall_accuracy': 0.8415841584158416}
			------------EPOCH 20---------------
Loss:  tensor(0.0106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0090, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1145, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0118, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0380, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1847, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1191, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1925, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0545, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0024, 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(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0883, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0147, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0264, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0612, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1225, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0609, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2194, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0090, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1544, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0145, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0432, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0059, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5437392795883362, 'recall': 0.6015180265654649, 'f1': 0.5711711711711712, 'number': 527}, 'P': {'precision': 0.6666666666666666, 'recall': 0.7083333333333334, 'f1': 0.6868686868686869, 'number': 912}, 'overall_precision': 0.6204896907216495, 'overall_recall': 0.6692147324530924, 'overall_f1': 0.6439317953861584, 'overall_accuracy': 0.8467034203420342}
			------------EPOCH 21---------------
Loss:  tensor(0.0023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1077, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0314, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1646, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1105, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2202, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0456, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0292, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0136, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0236, 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(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1201, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0051, 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(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0293, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0501, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0688, 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(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1955, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1721, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0630, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0057, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5599250936329588, 'recall': 0.5673624288425048, 'f1': 0.5636192271442035, 'number': 527}, 'P': {'precision': 0.6442687747035574, 'recall': 0.7149122807017544, 'f1': 0.6777546777546777, 'number': 912}, 'overall_precision': 0.6151358344113842, 'overall_recall': 0.6608756080611535, 'overall_f1': 0.6371859296482412, 'overall_accuracy': 0.8417810531053105}
			------------EPOCH 22---------------
Loss:  tensor(0.0021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0143, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1170, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0299, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1554, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0413, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0144, 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(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0907, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, 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(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0262, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0763, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0983, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0231, 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(0.0046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2636, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1770, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0174, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0430, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0054, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5274725274725275, 'recall': 0.5464895635673624, 'f1': 0.5368126747437093, 'number': 527}, 'P': {'precision': 0.6281553398058253, 'recall': 0.7094298245614035, 'f1': 0.666323377960865, 'number': 912}, 'overall_precision': 0.5932741116751269, 'overall_recall': 0.6497567755385685, 'overall_f1': 0.6202321724709784, 'overall_accuracy': 0.8413310081008101}
			------------EPOCH 23---------------
Loss:  tensor(0.0024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0999, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0284, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1837, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0170, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1769, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0488, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0166, 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(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0888, 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(0.0003, 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(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0180, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0437, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0549, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0195, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1828, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1324, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0142, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0658, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0049, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5422885572139303, 'recall': 0.6204933586337761, 'f1': 0.5787610619469026, 'number': 527}, 'P': {'precision': 0.6630785791173305, 'recall': 0.6754385964912281, 'f1': 0.6692015209125475, 'number': 912}, 'overall_precision': 0.6155352480417755, 'overall_recall': 0.655316191799861, 'overall_f1': 0.6348030966004713, 'overall_accuracy': 0.8395308280828083}
			------------EPOCH 24---------------
Loss:  tensor(0.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0895, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0322, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1305, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0228, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0930, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1183, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0355, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0141, 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(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0685, 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(0.0002, 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(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0361, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0398, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0658, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0112, 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(0.0038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1639, 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(0.1145, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0133, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0327, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0045, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5285234899328859, 'recall': 0.5977229601518027, 'f1': 0.5609973285841495, 'number': 527}, 'P': {'precision': 0.6541666666666667, 'recall': 0.6885964912280702, 'f1': 0.6709401709401709, 'number': 912}, 'overall_precision': 0.6060411311053985, 'overall_recall': 0.655316191799861, 'overall_f1': 0.6297161936560935, 'overall_accuracy': 0.8402340234023402}
			------------EPOCH 25---------------
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0666, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0374, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0170, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0370, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0077, 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(0.0160, 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(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0788, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, 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(0.0251, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0228, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0381, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0109, 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(0.0036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1338, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0327, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0044, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5597014925373134, 'recall': 0.5692599620493358, 'f1': 0.5644402634054563, 'number': 527}, 'P': {'precision': 0.6376237623762376, 'recall': 0.706140350877193, 'f1': 0.6701352757544224, 'number': 912}, 'overall_precision': 0.610608020698577, 'overall_recall': 0.6560111188325226, 'overall_f1': 0.6324958123953099, 'overall_accuracy': 0.8431593159315932}
			------------EPOCH 26---------------
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0652, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0348, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0163, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0853, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0084, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0147, 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(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0435, 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(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0239, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0444, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1734, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0541, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0041, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5449640287769785, 'recall': 0.5749525616698292, 'f1': 0.5595567867036011, 'number': 527}, 'P': {'precision': 0.6502016129032258, 'recall': 0.7072368421052632, 'f1': 0.6775210084033614, 'number': 912}, 'overall_precision': 0.6124031007751938, 'overall_recall': 0.6587908269631688, 'overall_f1': 0.6347505858721124, 'overall_accuracy': 0.8413310081008101}
			------------EPOCH 27---------------
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0407, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0292, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0757, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0140, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0829, 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(0.0063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0102, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0175, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0527, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.7942e-05, 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(0.0010, 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(0.0102, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0212, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0313, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1341, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0749, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0513, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0037, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5194174757281553, 'recall': 0.6091081593927894, 'f1': 0.5606986899563319, 'number': 527}, 'P': {'precision': 0.648109243697479, 'recall': 0.6765350877192983, 'f1': 0.6620171673819744, 'number': 912}, 'overall_precision': 0.597452229299363, 'overall_recall': 0.6518415566365532, 'overall_f1': 0.6234629444998339, 'overall_accuracy': 0.8404590459045904}
			------------EPOCH 28---------------
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0324, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0327, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0787, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0222, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0910, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0775, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0185, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0156, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0156, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0334, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.5680e-05, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, 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(0.0111, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0174, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0251, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0688, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0248, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0039, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5543478260869565, 'recall': 0.5806451612903226, 'f1': 0.5671918443002781, 'number': 527}, 'P': {'precision': 0.6488933601609658, 'recall': 0.7072368421052632, 'f1': 0.6768100734522561, 'number': 912}, 'overall_precision': 0.6151358344113842, 'overall_recall': 0.6608756080611535, 'overall_f1': 0.6371859296482412, 'overall_accuracy': 0.8436093609360936}
			------------EPOCH 29---------------
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0309, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0638, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0141, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1147, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0531, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, 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(0.0077, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0149, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0342, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.8906e-05, 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(0.0011, 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(0.0141, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0121, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0264, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1076, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0778, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0648, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0038, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5487179487179488, 'recall': 0.6091081593927894, 'f1': 0.5773381294964028, 'number': 527}, 'P': {'precision': 0.6556016597510373, 'recall': 0.6929824561403509, 'f1': 0.673773987206823, 'number': 912}, 'overall_precision': 0.6152356358941252, 'overall_recall': 0.6622654621264767, 'overall_f1': 0.6378848728246319, 'overall_accuracy': 0.8405996849684968}
			------------EPOCH 30---------------
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0251, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0430, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0489, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0135, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0938, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0563, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0165, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0166, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.7635e-05, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0358, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.5001e-05, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, 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(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0129, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0165, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0389, 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(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0711, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0418, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0034, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5133779264214047, 'recall': 0.5825426944971537, 'f1': 0.5457777777777778, 'number': 527}, 'P': {'precision': 0.64, 'recall': 0.6842105263157895, 'f1': 0.6613672496025438, 'number': 912}, 'overall_precision': 0.5918626827717737, 'overall_recall': 0.6469770674079222, 'overall_f1': 0.6181938911022578, 'overall_accuracy': 0.838996399639964}


		-------------RUN 3-----------
			------------EPOCH 1---------------
Loss:  tensor(3.6992, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6144, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2549, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9819, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6542, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1190, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6531, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5116, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1938, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5322, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4318, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4321, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1694, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2545, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2824, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2609, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1815, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2772, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2725, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2318, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7402, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2716, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2908, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0194, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9941, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8882, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8401, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3832, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2637, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7598, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7911, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3885, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6791, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6267, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7031, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.34594594594594597, 'recall': 0.11851851851851852, 'f1': 0.17655172413793105, 'number': 540}, 'P': {'precision': 0.45952575633687653, 'recall': 0.6386363636363637, 'f1': 0.5344745601521637, 'number': 880}, 'overall_precision': 0.4446022727272727, 'overall_recall': 0.44084507042253523, 'overall_f1': 0.4427157001414427, 'overall_accuracy': 0.7385660949714152}
			------------EPOCH 2---------------
Loss:  tensor(1.7569, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6853, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4465, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8742, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0204, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4210, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4378, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1305, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4897, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4502, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2816, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3963, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3847, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1668, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0808, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2637, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3442, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4653, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4861, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1686, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1840, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1509, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2623, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1672, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1571, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9340, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2550, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4812, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2981, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1474, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1871, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0586, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0935, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.47246376811594204, 'recall': 0.30185185185185187, 'f1': 0.36836158192090396, 'number': 540}, 'P': {'precision': 0.44124336618650495, 'recall': 0.6613636363636364, 'f1': 0.529331514324693, 'number': 880}, 'overall_precision': 0.44771634615384615, 'overall_recall': 0.5246478873239436, 'overall_f1': 0.48313878080415046, 'overall_accuracy': 0.7707385369268464}
			------------EPOCH 3---------------
Loss:  tensor(1.3125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1837, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0194, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4076, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7241, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9752, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3137, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0826, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0689, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0930, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2143, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1947, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0615, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1218, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8333, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0610, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0258, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0120, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3577, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0956, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8972, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8249, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6646, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3954, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3541, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8925, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9127, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.479933110367893, 'recall': 0.5314814814814814, 'f1': 0.5043936731107206, 'number': 540}, 'P': {'precision': 0.5338645418326693, 'recall': 0.6090909090909091, 'f1': 0.5690021231422505, 'number': 880}, 'overall_precision': 0.5137328339575531, 'overall_recall': 0.5795774647887324, 'overall_f1': 0.5446724023825282, 'overall_accuracy': 0.821374402053436}
			------------EPOCH 4---------------
Loss:  tensor(0.9413, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6959, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1736, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6439, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7169, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5592, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9430, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2466, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7483, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0476, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8696, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7771, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1915, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0755, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9166, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9197, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5419, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0378, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7144, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6751, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7114, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2679, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0717, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0960, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7744, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4926, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2806, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2603, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1518, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5601, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7394, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5515789473684211, 'recall': 0.48518518518518516, 'f1': 0.5162561576354681, 'number': 540}, 'P': {'precision': 0.5700667302192565, 'recall': 0.6795454545454546, 'f1': 0.6200103680663557, 'number': 880}, 'overall_precision': 0.5643044619422573, 'overall_recall': 0.6056338028169014, 'overall_f1': 0.5842391304347825, 'overall_accuracy': 0.8181950764204877}
			------------EPOCH 5---------------
Loss:  tensor(0.7633, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5815, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6465, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5395, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4212, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4889, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3645, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8237, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1897, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5597, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0233, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6257, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6450, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1202, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8898, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6936, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6572, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3543, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0366, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5578, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4875, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1952, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7204, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0841, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5423, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3903, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4925, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3966, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3374, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1795, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2392, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2625, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4340, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6825, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4766, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4618, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.62751677852349, 'recall': 0.34629629629629627, 'f1': 0.44630071599045346, 'number': 540}, 'P': {'precision': 0.4939301972685888, 'recall': 0.7397727272727272, 'f1': 0.5923566878980893, 'number': 880}, 'overall_precision': 0.5185643564356436, 'overall_recall': 0.5901408450704225, 'overall_f1': 0.5520421607378129, 'overall_accuracy': 0.795764788239412}
			------------EPOCH 6---------------
Loss:  tensor(0.8597, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5282, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6659, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4415, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8539, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3277, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3481, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7561, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2432, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6341, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0228, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7223, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5792, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0841, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8114, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5331, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5239, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0196, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6630, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4814, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6391, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2218, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7524, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0482, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0674, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3786, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2897, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3454, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3177, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2691, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1513, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1978, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1361, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5333, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6531, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5786, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6236, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5457875457875457, 'recall': 0.5518518518518518, 'f1': 0.5488029465930019, 'number': 540}, 'P': {'precision': 0.629746835443038, 'recall': 0.678409090909091, 'f1': 0.6531728665207878, 'number': 880}, 'overall_precision': 0.5990629183400268, 'overall_recall': 0.6302816901408451, 'overall_f1': 0.6142759094028827, 'overall_accuracy': 0.830154007700385}
			------------EPOCH 7---------------
Loss:  tensor(0.4468, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2443, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8169, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5120, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3782, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7603, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2412, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4292, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0288, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4314, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3678, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0601, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6160, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5663, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4943, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2991, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0238, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3877, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2746, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1418, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4841, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0187, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0403, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2413, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2140, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2335, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2497, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2738, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1609, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1554, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0942, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0978, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2434, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1742, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2921, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4917293233082707, 'recall': 0.6055555555555555, 'f1': 0.5427385892116182, 'number': 540}, 'P': {'precision': 0.6424242424242425, 'recall': 0.6022727272727273, 'f1': 0.621700879765396, 'number': 880}, 'overall_precision': 0.5751677852348993, 'overall_recall': 0.6035211267605634, 'overall_f1': 0.5890034364261169, 'overall_accuracy': 0.8194493057986233}
			------------EPOCH 8---------------
Loss:  tensor(0.3168, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2924, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2575, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2733, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6851, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1307, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1407, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3956, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1461, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2506, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0111, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2650, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2522, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0329, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3840, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3725, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0136, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1872, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1227, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1605, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0650, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4857, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0111, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0177, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1486, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1501, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1307, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1174, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0585, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0823, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0780, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1848, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2246, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1259, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1944, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5264054514480409, 'recall': 0.5722222222222222, 'f1': 0.5483584738243122, 'number': 540}, 'P': {'precision': 0.6459489456159823, 'recall': 0.6613636363636364, 'f1': 0.6535654126895003, 'number': 880}, 'overall_precision': 0.5987903225806451, 'overall_recall': 0.6274647887323944, 'overall_f1': 0.6127922971114168, 'overall_accuracy': 0.8257204526893012}
			------------EPOCH 9---------------
Loss:  tensor(0.3055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1581, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1243, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5255, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0771, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0764, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2105, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2447, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1496, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0239, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3952, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2922, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2768, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1296, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0837, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0651, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0943, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0700, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0682, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0868, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1504, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0813, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0452, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0493, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0545, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1348, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0557, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1158, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.47583081570996977, 'recall': 0.5833333333333334, 'f1': 0.524126455906822, 'number': 540}, 'P': {'precision': 0.6556818181818181, 'recall': 0.6556818181818181, 'f1': 0.6556818181818181, 'number': 880}, 'overall_precision': 0.5784695201037614, 'overall_recall': 0.6281690140845071, 'overall_f1': 0.6022957461174883, 'overall_accuracy': 0.8092404620231012}
			------------EPOCH 10---------------
Loss:  tensor(0.2308, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1270, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4726, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0616, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0708, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0834, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2418, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0924, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1531, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1084, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0140, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3529, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2665, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2353, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0822, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0937, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0674, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0550, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0499, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3790, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0512, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0414, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0721, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0868, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0531, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0344, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0548, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0571, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1174, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1187, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0455, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0843, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.487256371814093, 'recall': 0.6018518518518519, 'f1': 0.5385252692626347, 'number': 540}, 'P': {'precision': 0.6180021953896817, 'recall': 0.6397727272727273, 'f1': 0.6286990508096035, 'number': 880}, 'overall_precision': 0.5627376425855514, 'overall_recall': 0.6253521126760564, 'overall_f1': 0.5923949299533022, 'overall_accuracy': 0.8188951114222378}
			------------EPOCH 11---------------
Loss:  tensor(0.2515, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0501, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0977, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0415, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4549, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0753, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0694, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0503, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0856, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0498, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1105, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1211, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3751, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2392, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2261, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0481, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0751, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0609, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0320, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3890, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0578, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0539, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1318, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0823, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0183, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0691, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0667, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1688, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0305, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0896, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.505, 'recall': 0.5611111111111111, 'f1': 0.531578947368421, 'number': 540}, 'P': {'precision': 0.5954825462012321, 'recall': 0.6590909090909091, 'f1': 0.6256742179072277, 'number': 880}, 'overall_precision': 0.5609911054637865, 'overall_recall': 0.6218309859154929, 'overall_f1': 0.5898463593854374, 'overall_accuracy': 0.8271788589429472}
			------------EPOCH 12---------------
Loss:  tensor(0.1780, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0429, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1130, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0325, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4487, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0518, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0485, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0621, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1631, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0395, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3173, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2358, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2409, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0572, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0736, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0555, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0442, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0102, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2727, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0505, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0834, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0793, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0571, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0301, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0356, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0444, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0974, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0899, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0400, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0779, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5560165975103735, 'recall': 0.4962962962962963, 'f1': 0.5244618395303327, 'number': 540}, 'P': {'precision': 0.582089552238806, 'recall': 0.7090909090909091, 'f1': 0.6393442622950819, 'number': 880}, 'overall_precision': 0.574002574002574, 'overall_recall': 0.6281690140845071, 'overall_f1': 0.5998655010087425, 'overall_accuracy': 0.8209368801773422}
			------------EPOCH 13---------------
Loss:  tensor(0.2502, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0422, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0979, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0792, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3535, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0465, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0254, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0544, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1332, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0565, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0792, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0152, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2975, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1736, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0446, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0501, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0214, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0613, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0107, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0706, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0644, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0282, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0539, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0966, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0254, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0444, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0476, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1555, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1573, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0283, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0798, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5699588477366255, 'recall': 0.512962962962963, 'f1': 0.5399610136452241, 'number': 540}, 'P': {'precision': 0.648109243697479, 'recall': 0.7011363636363637, 'f1': 0.673580786026201, 'number': 880}, 'overall_precision': 0.6216968011126565, 'overall_recall': 0.6295774647887324, 'overall_f1': 0.6256123163051085, 'overall_accuracy': 0.8289872826974682}
			------------EPOCH 14---------------
Loss:  tensor(0.1647, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0285, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0816, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3540, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0301, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0273, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0490, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0649, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0674, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1437, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0859, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2813, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2257, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1783, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0897, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0670, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0492, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0341, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0076, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2441, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0432, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0346, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0211, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0516, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0692, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0356, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0190, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0444, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0623, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0618, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0500, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0662, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5642570281124498, 'recall': 0.5203703703703704, 'f1': 0.5414258188824662, 'number': 540}, 'P': {'precision': 0.6068204613841525, 'recall': 0.6875, 'f1': 0.6446457112413425, 'number': 880}, 'overall_precision': 0.5926421404682274, 'overall_recall': 0.6239436619718309, 'overall_f1': 0.6078902229845626, 'overall_accuracy': 0.818807607047019}
			------------EPOCH 15---------------
Loss:  tensor(0.1868, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0314, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0741, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0480, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2996, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0239, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0260, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0398, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0954, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0694, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0286, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0838, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0565, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2392, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0239, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0341, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0247, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0506, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0270, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0297, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0574, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0376, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0197, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0655, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0776, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0212, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0495, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5766871165644172, 'recall': 0.5222222222222223, 'f1': 0.5481049562682215, 'number': 540}, 'P': {'precision': 0.5819366852886406, 'recall': 0.7102272727272727, 'f1': 0.63971340839304, 'number': 880}, 'overall_precision': 0.5802943058221369, 'overall_recall': 0.6387323943661972, 'overall_f1': 0.6081126382836072, 'overall_accuracy': 0.8258662933146658}
			------------EPOCH 16---------------
Loss:  tensor(0.1287, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0731, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0166, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0126, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0340, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0934, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0957, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0128, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0377, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0423, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2203, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1353, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0395, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0254, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2485, 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(0.0239, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0218, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0643, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0516, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0323, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0376, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0538, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0114, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0339, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5290806754221389, 'recall': 0.5222222222222223, 'f1': 0.5256290773532152, 'number': 540}, 'P': {'precision': 0.5297346752058555, 'recall': 0.6579545454545455, 'f1': 0.5869234668018247, 'number': 880}, 'overall_precision': 0.5295202952029521, 'overall_recall': 0.606338028169014, 'overall_f1': 0.5653315824031517, 'overall_accuracy': 0.8219869326799674}
			------------EPOCH 17---------------
Loss:  tensor(0.1177, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0182, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0442, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0204, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2518, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0218, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0345, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0774, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0660, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0258, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0481, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2880, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1386, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0325, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0314, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0248, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2588, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0292, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0090, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0214, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0392, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0501, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0637, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0499, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0557, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0105, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0309, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4821705426356589, 'recall': 0.575925925925926, 'f1': 0.5248945147679325, 'number': 540}, 'P': {'precision': 0.575531914893617, 'recall': 0.6147727272727272, 'f1': 0.5945054945054945, 'number': 880}, 'overall_precision': 0.5375394321766561, 'overall_recall': 0.6, 'overall_f1': 0.5670549084858568, 'overall_accuracy': 0.8194493057986233}
			------------EPOCH 18---------------
Loss:  tensor(0.1302, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0413, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0123, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4077, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0153, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0361, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0862, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0900, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0268, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2441, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1090, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1170, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0123, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0076, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0190, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2632, 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(0.0320, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0236, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0341, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0748, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0476, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0170, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0429, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0325, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0420, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0187, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5161290322580645, 'recall': 0.562962962962963, 'f1': 0.5385296722763508, 'number': 540}, 'P': {'precision': 0.6202804746494067, 'recall': 0.6534090909090909, 'f1': 0.6364139457664637, 'number': 880}, 'overall_precision': 0.5798153034300791, 'overall_recall': 0.6190140845070422, 'overall_f1': 0.5987738419618528, 'overall_accuracy': 0.8206160308015401}
			------------EPOCH 19---------------
Loss:  tensor(0.1467, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0296, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2791, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0211, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0303, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0645, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0626, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0339, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0320, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0547, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2317, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1137, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0168, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0090, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1730, 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(0.0110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0212, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0413, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0210, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0373, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0497, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0572, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0195, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5664621676891616, 'recall': 0.512962962962963, 'f1': 0.5383867832847424, 'number': 540}, 'P': {'precision': 0.6479750778816199, 'recall': 0.7090909090909091, 'f1': 0.6771568095496473, 'number': 880}, 'overall_precision': 0.6205234159779615, 'overall_recall': 0.6345070422535212, 'overall_f1': 0.6274373259052926, 'overall_accuracy': 0.823591179558978}
			------------EPOCH 20---------------
Loss:  tensor(0.1160, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0390, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2600, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0200, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0361, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0506, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0427, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0306, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0269, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1688, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0788, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0909, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0112, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0143, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1809, 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(0.0127, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0198, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0259, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0343, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0118, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0160, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0194, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5574712643678161, 'recall': 0.5388888888888889, 'f1': 0.5480225988700566, 'number': 540}, 'P': {'precision': 0.6223350253807106, 'recall': 0.696590909090909, 'f1': 0.6573726541554958, 'number': 880}, 'overall_precision': 0.5998672859986729, 'overall_recall': 0.6366197183098592, 'overall_f1': 0.6176973009907756, 'overall_accuracy': 0.8277038851942597}
			------------EPOCH 21---------------
Loss:  tensor(0.0846, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2112, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0359, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0436, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0329, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0126, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0244, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1516, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0643, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0712, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1476, 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(0.0171, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0194, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0325, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0305, 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(0.0212, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0163, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0176, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0154, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5444646098003629, 'recall': 0.5555555555555556, 'f1': 0.5499541704857929, 'number': 540}, 'P': {'precision': 0.6266944734098019, 'recall': 0.6829545454545455, 'f1': 0.6536160957041872, 'number': 880}, 'overall_precision': 0.5966887417218543, 'overall_recall': 0.6345070422535212, 'overall_f1': 0.6150170648464164, 'overall_accuracy': 0.82650799206627}
			------------EPOCH 22---------------
Loss:  tensor(0.1094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0144, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1873, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0279, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0367, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0340, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0115, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0225, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1465, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0564, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0607, 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(0.0100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1212, 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(0.0045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0195, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0324, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0210, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0170, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0137, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0132, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5533596837944664, 'recall': 0.5185185185185185, 'f1': 0.5353728489483748, 'number': 540}, 'P': {'precision': 0.6511134676564156, 'recall': 0.6977272727272728, 'f1': 0.673614920460779, 'number': 880}, 'overall_precision': 0.6169772256728778, 'overall_recall': 0.6295774647887324, 'overall_f1': 0.6232136632973162, 'overall_accuracy': 0.8283164158207911}
			------------EPOCH 23---------------
Loss:  tensor(0.0981, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0167, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2164, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0337, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0259, 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(0.0083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1307, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0480, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0497, 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(0.0042, 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(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1209, 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(0.0038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0163, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0396, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0316, 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(0.0240, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0170, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0198, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0184, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5444444444444444, 'recall': 0.5444444444444444, 'f1': 0.5444444444444444, 'number': 540}, 'P': {'precision': 0.6192946058091287, 'recall': 0.678409090909091, 'f1': 0.6475054229934925, 'number': 880}, 'overall_precision': 0.5924202127659575, 'overall_recall': 0.6274647887323944, 'overall_f1': 0.6094391244870042, 'overall_accuracy': 0.8260413020651033}
			------------EPOCH 24---------------
Loss:  tensor(0.0813, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0292, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1750, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0396, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0290, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0224, 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(0.0076, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0188, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1413, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0609, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0593, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0015, 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>)
Loss:  tensor(0.1273, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0042, 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(0.0199, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0286, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0272, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0230, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0129, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5708061002178649, 'recall': 0.48518518518518516, 'f1': 0.5245245245245245, 'number': 540}, 'P': {'precision': 0.5831012070566388, 'recall': 0.7136363636363636, 'f1': 0.6417986714358712, 'number': 880}, 'overall_precision': 0.5794270833333334, 'overall_recall': 0.6267605633802817, 'overall_f1': 0.6021650879566983, 'overall_accuracy': 0.8219285964298215}
			------------EPOCH 25---------------
Loss:  tensor(0.1377, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0234, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1972, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0231, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0458, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0496, 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(0.0071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0208, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1319, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0476, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0601, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0055, 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(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1709, 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(0.0030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0193, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0303, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0698, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0090, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0348, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0337, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0527, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0146, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.540952380952381, 'recall': 0.5259259259259259, 'f1': 0.5333333333333334, 'number': 540}, 'P': {'precision': 0.6517571884984026, 'recall': 0.6954545454545454, 'f1': 0.6728971962616823, 'number': 880}, 'overall_precision': 0.6120218579234973, 'overall_recall': 0.6309859154929578, 'overall_f1': 0.6213592233009709, 'overall_accuracy': 0.8303873527009684}
			------------EPOCH 26---------------
Loss:  tensor(0.0819, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1513, 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(0.0379, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0406, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0268, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0272, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1441, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0653, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0670, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1147, 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(0.0021, 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(0.0151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0408, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0254, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0188, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0199, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0148, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.502495840266223, 'recall': 0.5592592592592592, 'f1': 0.5293602103418054, 'number': 540}, 'P': {'precision': 0.6294691224268689, 'recall': 0.6602272727272728, 'f1': 0.644481419855796, 'number': 880}, 'overall_precision': 0.5793963254593176, 'overall_recall': 0.6218309859154929, 'overall_f1': 0.5998641304347825, 'overall_accuracy': 0.8275288764438222}
			------------EPOCH 27---------------
Loss:  tensor(0.0902, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0123, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0289, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0573, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0512, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1210, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0453, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0467, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, 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(0.1102, 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(0.0021, 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(0.0164, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0268, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0363, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0188, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0218, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0254, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0187, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.579520697167756, 'recall': 0.4925925925925926, 'f1': 0.5325325325325326, 'number': 540}, 'P': {'precision': 0.6113837095191365, 'recall': 0.7079545454545455, 'f1': 0.6561348077935757, 'number': 880}, 'overall_precision': 0.6014884979702301, 'overall_recall': 0.6260563380281691, 'overall_f1': 0.6135265700483092, 'overall_accuracy': 0.8229786489324467}
			------------EPOCH 28---------------
Loss:  tensor(0.1004, 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(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1623, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0284, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0395, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0219, 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(0.0046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0166, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0996, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0435, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0411, 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(0.0027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0935, 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(0.0017, 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(0.0218, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0380, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0139, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0112, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0154, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0125, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5504761904761905, 'recall': 0.5351851851851852, 'f1': 0.5427230046948357, 'number': 540}, 'P': {'precision': 0.6203703703703703, 'recall': 0.6852272727272727, 'f1': 0.6511879049676026, 'number': 880}, 'overall_precision': 0.5958583834335337, 'overall_recall': 0.6281690140845071, 'overall_f1': 0.6115872471717518, 'overall_accuracy': 0.8285789289464474}
			------------EPOCH 29---------------
Loss:  tensor(0.0848, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1208, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0355, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0272, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0142, 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>)
Loss:  tensor(0.0038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0878, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0277, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0268, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0046, 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(0.0010, 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(0.0656, 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(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0164, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0148, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0132, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0092, 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>)
				 {'C': {'precision': 0.5583657587548638, 'recall': 0.5314814814814814, 'f1': 0.5445920303605313, 'number': 540}, 'P': {'precision': 0.6332644628099173, 'recall': 0.696590909090909, 'f1': 0.6634199134199135, 'number': 880}, 'overall_precision': 0.6072874493927125, 'overall_recall': 0.6338028169014085, 'overall_f1': 0.6202618883528601, 'overall_accuracy': 0.8311165558277914}
			------------EPOCH 30---------------
Loss:  tensor(0.1016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0993, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0258, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0287, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0132, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0102, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0655, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0214, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0053, 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(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0625, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, 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(0.0141, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0121, 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(0.0094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0146, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5393053016453382, 'recall': 0.5462962962962963, 'f1': 0.5427782888684451, 'number': 540}, 'P': {'precision': 0.6212278876170656, 'recall': 0.678409090909091, 'f1': 0.6485605649103748, 'number': 880}, 'overall_precision': 0.5915119363395226, 'overall_recall': 0.6281690140845071, 'overall_f1': 0.6092896174863389, 'overall_accuracy': 0.8289872826974682}


		-------------RUN 4-----------
			------------EPOCH 1---------------
Loss:  tensor(0.3259, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4545, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2682, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0553, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9785, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5841, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2398, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6570, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5837, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6659, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1956, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8231, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0239, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4707, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3182, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2909, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2480, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6185, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2898, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2748, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1114, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2807, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3807, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8499, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8443, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3807, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8441, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2985, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1567, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9214, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.24018475750577367, 'recall': 0.23853211009174313, 'f1': 0.23935558112773303, 'number': 436}, 'P': {'precision': 0.3168103448275862, 'recall': 0.3941018766756032, 'f1': 0.35125448028673834, 'number': 746}, 'overall_precision': 0.2924320352681852, 'overall_recall': 0.33671742808798644, 'overall_f1': 0.31301612268973655, 'overall_accuracy': 0.717059829059829}
			------------EPOCH 2---------------
Loss:  tensor(0.0935, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7549, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6482, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1649, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5380, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7310, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3530, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0892, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3678, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6961, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4614, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5858, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4512, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0414, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6207, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3116, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2208, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4867, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9596, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4345, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3568, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3231, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3628, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1572, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4766, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1605, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3529, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9432, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2279, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2916, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1112, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0896, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4713, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3781362007168459, 'recall': 0.48394495412844035, 'f1': 0.4245472837022133, 'number': 436}, 'P': {'precision': 0.5315186246418339, 'recall': 0.4973190348525469, 'f1': 0.5138504155124655, 'number': 746}, 'overall_precision': 0.46337579617834396, 'overall_recall': 0.49238578680203043, 'overall_f1': 0.4774405250205086, 'overall_accuracy': 0.777982905982906}
			------------EPOCH 3---------------
Loss:  tensor(0.0507, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2674, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0552, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0464, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2204, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8481, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7107, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0130, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0900, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9646, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0500, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0607, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7949, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9962, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0186, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9363, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1413, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0674, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9394, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9254, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9387, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0960, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5811, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8495, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2988, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6309, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8415, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2554, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7237, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7625, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7986, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0603, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1363, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.42857142857142855, 'recall': 0.536697247706422, 'f1': 0.4765784114052953, 'number': 436}, 'P': {'precision': 0.5085995085995086, 'recall': 0.5549597855227882, 'f1': 0.5307692307692308, 'number': 746}, 'overall_precision': 0.4764705882352941, 'overall_recall': 0.5482233502538071, 'overall_f1': 0.5098347757671124, 'overall_accuracy': 0.8004102564102564}
			------------EPOCH 4---------------
Loss:  tensor(0.0359, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9294, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7490, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4772, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7879, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8809, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5499, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5269, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6806, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7968, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6320, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7968, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8655, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6407, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6991, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6656, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8531, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6636, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0891, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7707, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7520, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6270, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6979, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6251, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5717, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3858, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2495, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6677, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2631, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4597, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5746, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2632, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4571, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4156, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5514, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0419, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1097, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.44310344827586207, 'recall': 0.5894495412844036, 'f1': 0.5059055118110236, 'number': 436}, 'P': {'precision': 0.5826558265582655, 'recall': 0.5764075067024129, 'f1': 0.5795148247978437, 'number': 746}, 'overall_precision': 0.5212443095599393, 'overall_recall': 0.5812182741116751, 'overall_f1': 0.5496, 'overall_accuracy': 0.8034871794871795}
			------------EPOCH 5---------------
Loss:  tensor(0.0221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7625, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4925, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2724, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5372, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7128, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3173, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5946, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4728, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5742, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6567, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5211, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4594, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2414, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4745, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7482, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5551, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0589, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6950, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3314, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5512, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3850, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4397, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3769, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5686, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0644, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5242, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2137, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3545, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4128, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3415, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3654, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4779, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0195, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9688, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.49671772428884026, 'recall': 0.5206422018348624, 'f1': 0.5083986562150057, 'number': 436}, 'P': {'precision': 0.6190476190476191, 'recall': 0.6447721179624665, 'f1': 0.6316480630334865, 'number': 746}, 'overall_precision': 0.5737439222042139, 'overall_recall': 0.5989847715736041, 'overall_f1': 0.5860927152317881, 'overall_accuracy': 0.8065641025641026}
			------------EPOCH 6---------------
Loss:  tensor(0.0181, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5559, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3288, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4503, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6722, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2965, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2208, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3575, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4485, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3557, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3271, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6949, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4700, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4090, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1838, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6418, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7792, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3326, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0656, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4252, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2193, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4413, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2815, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4448, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3942, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0420, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2414, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1881, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2852, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2946, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2687, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3425, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1947, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2730, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2927, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0159, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8493, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.43149284253578735, 'recall': 0.48394495412844035, 'f1': 0.4562162162162162, 'number': 436}, 'P': {'precision': 0.5949868073878628, 'recall': 0.6045576407506702, 'f1': 0.5997340425531915, 'number': 746}, 'overall_precision': 0.5308740978348035, 'overall_recall': 0.560067681895093, 'overall_f1': 0.5450802799505969, 'overall_accuracy': 0.7993846153846154}
			------------EPOCH 7---------------
Loss:  tensor(0.0147, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6170, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2851, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2330, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5342, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1479, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1946, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2988, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1772, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2190, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5150, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3455, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0994, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2243, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4603, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1526, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0271, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2940, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1435, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4571, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2677, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1633, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1902, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1831, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1950, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1362, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1799, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2265, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2974, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1188, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1580, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1381, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1160, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7526, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5828402366863905, 'recall': 0.4518348623853211, 'f1': 0.5090439276485788, 'number': 436}, 'P': {'precision': 0.5768811341330425, 'recall': 0.7091152815013405, 'f1': 0.6361996392062538, 'number': 746}, 'overall_precision': 0.5784860557768924, 'overall_recall': 0.6142131979695431, 'overall_f1': 0.5958145260566269, 'overall_accuracy': 0.8291623931623932}
			------------EPOCH 8---------------
Loss:  tensor(0.0117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5360, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2971, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1262, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1620, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1516, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1960, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1328, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5306, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3529, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1500, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0733, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1760, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3704, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0947, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2650, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4606, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2458, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1629, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1552, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0805, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0750, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0802, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1418, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1602, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1790, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0873, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1523, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2359, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0076, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6625, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5125523012552301, 'recall': 0.5619266055045872, 'f1': 0.5361050328227571, 'number': 436}, 'P': {'precision': 0.6167741935483871, 'recall': 0.6407506702412868, 'f1': 0.62853385930309, 'number': 746}, 'overall_precision': 0.5770151636073424, 'overall_recall': 0.6116751269035533, 'overall_f1': 0.5938398357289528, 'overall_accuracy': 0.8134358974358974}
			------------EPOCH 9---------------
Loss:  tensor(0.0051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3346, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0638, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1279, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0922, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1570, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0955, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4121, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1183, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0595, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1525, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3403, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0801, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0077, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1863, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0865, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2277, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1463, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0702, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0605, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0717, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1688, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1342, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1563, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0866, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0972, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0660, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5413, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5300925925925926, 'recall': 0.5252293577981652, 'f1': 0.5276497695852533, 'number': 436}, 'P': {'precision': 0.6007025761124122, 'recall': 0.6876675603217158, 'f1': 0.64125, 'number': 746}, 'overall_precision': 0.5769828926905132, 'overall_recall': 0.6277495769881557, 'overall_f1': 0.6012965964343598, 'overall_accuracy': 0.8249230769230769}
			------------EPOCH 10---------------
Loss:  tensor(0.0033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2635, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0755, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0449, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0971, 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(0.0756, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0580, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1355, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0814, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0368, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3368, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2382, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0824, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0240, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1234, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3173, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0444, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0554, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0618, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0795, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0935, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0687, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0383, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0370, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1557, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0959, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1308, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0745, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0806, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0566, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0479, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6023, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.49478079331941544, 'recall': 0.5435779816513762, 'f1': 0.5180327868852458, 'number': 436}, 'P': {'precision': 0.6078914919852034, 'recall': 0.660857908847185, 'f1': 0.6332691072575466, 'number': 746}, 'overall_precision': 0.5658914728682171, 'overall_recall': 0.6175972927241963, 'overall_f1': 0.5906148867313916, 'overall_accuracy': 0.8179487179487179}
			------------EPOCH 11---------------
Loss:  tensor(0.0051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2680, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0683, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0290, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0727, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2911, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0587, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0770, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0372, 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(0.0258, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3361, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2199, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0500, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0204, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1348, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3289, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0503, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0540, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1733, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0598, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0599, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0579, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0358, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0423, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0369, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0166, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0866, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0842, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0468, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0459, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0538, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0277, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4630, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.45036764705882354, 'recall': 0.5619266055045872, 'f1': 0.5, 'number': 436}, 'P': {'precision': 0.6055276381909548, 'recall': 0.646112600536193, 'f1': 0.6251621271076523, 'number': 746}, 'overall_precision': 0.5425373134328358, 'overall_recall': 0.6150592216582065, 'overall_f1': 0.5765265662172878, 'overall_accuracy': 0.8111111111111111}
			------------EPOCH 12---------------
Loss:  tensor(0.0039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2370, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0435, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0415, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1810, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0464, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0371, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0939, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0282, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0201, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3282, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0290, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0763, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2559, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0212, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0697, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0289, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1946, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0257, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0506, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0413, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0772, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0546, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0224, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1459, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0559, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1561, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0273, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0372, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0359, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4258, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4500818330605565, 'recall': 0.6307339449541285, 'f1': 0.5253104106972302, 'number': 436}, 'P': {'precision': 0.632088520055325, 'recall': 0.6126005361930295, 'f1': 0.6221919673247106, 'number': 746}, 'overall_precision': 0.5487256371814093, 'overall_recall': 0.6192893401015228, 'overall_f1': 0.5818759936406994, 'overall_accuracy': 0.8004786324786325}
			------------EPOCH 13---------------
Loss:  tensor(0.0029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1827, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0445, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0208, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1239, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1557, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0453, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0283, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0831, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0282, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0176, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3448, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2186, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0258, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0755, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2788, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0621, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1501, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0350, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1757, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0368, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0845, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0277, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0501, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0323, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0251, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0992, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0558, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0711, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0992, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0734, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0724, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4595, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.501984126984127, 'recall': 0.5802752293577982, 'f1': 0.5382978723404256, 'number': 436}, 'P': {'precision': 0.5977722772277227, 'recall': 0.6474530831099196, 'f1': 0.6216216216216215, 'number': 746}, 'overall_precision': 0.5609756097560976, 'overall_recall': 0.622673434856176, 'overall_f1': 0.5902165196471532, 'overall_accuracy': 0.8136068376068376}
			------------EPOCH 14---------------
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2154, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0756, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0437, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1686, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0352, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0400, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0345, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0826, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0750, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0301, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3085, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1924, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0832, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2830, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1686, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0894, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0234, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1828, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0503, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0458, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0506, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0979, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0911, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0821, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0744, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0504, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0312, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0494, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0418, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3816, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4567669172932331, 'recall': 0.5573394495412844, 'f1': 0.5020661157024794, 'number': 436}, 'P': {'precision': 0.5665488810365136, 'recall': 0.6447721179624665, 'f1': 0.6031347962382445, 'number': 746}, 'overall_precision': 0.5242577842143374, 'overall_recall': 0.6125211505922166, 'overall_f1': 0.5649629340616465, 'overall_accuracy': 0.8226324786324787}
			------------EPOCH 15---------------
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1647, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0482, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0105, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0296, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0345, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0257, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0653, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0489, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1663, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0717, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0736, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0283, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1737, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0373, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0369, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0731, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0155, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0844, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0546, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0695, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0419, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0585, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0504, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0384, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3778, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.44642857142857145, 'recall': 0.5160550458715596, 'f1': 0.4787234042553192, 'number': 436}, 'P': {'precision': 0.5378440366972477, 'recall': 0.628686327077748, 'f1': 0.5797280593325093, 'number': 746}, 'overall_precision': 0.5043604651162791, 'overall_recall': 0.5871404399323181, 'overall_f1': 0.5426114151681, 'overall_accuracy': 0.8161025641025641}
			------------EPOCH 16---------------
Loss:  tensor(0.0055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1474, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0434, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0152, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0524, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0819, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0319, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0282, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0400, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0485, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0210, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2228, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0392, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1813, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0220, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0876, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0269, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0801, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0324, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0309, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0393, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0105, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0983, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0972, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0566, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0440, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0383, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3250, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.455078125, 'recall': 0.5344036697247706, 'f1': 0.49156118143459915, 'number': 436}, 'P': {'precision': 0.49318181818181817, 'recall': 0.5817694369973191, 'f1': 0.5338253382533824, 'number': 746}, 'overall_precision': 0.4791666666666667, 'overall_recall': 0.5642978003384095, 'overall_f1': 0.5182595182595183, 'overall_accuracy': 0.811965811965812}
			------------EPOCH 17---------------
Loss:  tensor(0.0031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0317, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0191, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0343, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0750, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0346, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0285, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0244, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0453, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0257, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0128, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2272, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1454, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0132, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0275, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1766, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0392, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0157, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1659, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0409, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0741, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0326, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0750, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0468, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0208, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0121, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0878, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0531, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0531, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0450, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0126, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0273, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3625, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.46586345381526106, 'recall': 0.5321100917431193, 'f1': 0.4967880085653105, 'number': 436}, 'P': {'precision': 0.5107758620689655, 'recall': 0.6353887399463807, 'f1': 0.5663082437275986, 'number': 746}, 'overall_precision': 0.4950911640953717, 'overall_recall': 0.5972927241962775, 'overall_f1': 0.5414110429447853, 'overall_accuracy': 0.8215042735042735}
			------------EPOCH 18---------------
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1334, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0308, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0738, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0415, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0271, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0301, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0706, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0400, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0336, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2374, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1398, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0315, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1909, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0102, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0343, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0175, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0337, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0646, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0498, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0963, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0642, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0443, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0553, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3395, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5651041666666666, 'recall': 0.49770642201834864, 'f1': 0.5292682926829269, 'number': 436}, 'P': {'precision': 0.592391304347826, 'recall': 0.7305630026809652, 'f1': 0.6542617046818727, 'number': 746}, 'overall_precision': 0.5843558282208589, 'overall_recall': 0.6446700507614214, 'overall_f1': 0.6130329847144006, 'overall_accuracy': 0.8314188034188034}
			------------EPOCH 19---------------
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1319, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0754, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0202, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0449, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0202, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0181, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1384, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0244, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0121, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0417, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2180, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0456, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0234, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1465, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0132, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0140, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0144, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0364, 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(0.1413, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0774, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0748, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0225, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3141, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.6171617161716172, 'recall': 0.4288990825688073, 'f1': 0.5060893098782139, 'number': 436}, 'P': {'precision': 0.6180400890868597, 'recall': 0.7439678284182306, 'f1': 0.6751824817518248, 'number': 746}, 'overall_precision': 0.6178184845961698, 'overall_recall': 0.6277495769881557, 'overall_f1': 0.6227444397817876, 'overall_accuracy': 0.8322393162393162}
			------------EPOCH 20---------------
Loss:  tensor(0.0055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1582, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0608, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0238, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0331, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1757, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0393, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0548, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1299, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0295, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1834, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0160, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0345, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1436, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0212, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0648, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0285, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0201, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0430, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0258, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0988, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0357, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0712, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0542, 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(0.0092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3192, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5463414634146342, 'recall': 0.5137614678899083, 'f1': 0.5295508274231678, 'number': 436}, 'P': {'precision': 0.5741525423728814, 'recall': 0.7265415549597856, 'f1': 0.6414201183431953, 'number': 746}, 'overall_precision': 0.5657311669128509, 'overall_recall': 0.6480541455160744, 'overall_f1': 0.6041009463722398, 'overall_accuracy': 0.8281367521367521}
			------------EPOCH 21---------------
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1222, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0147, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0115, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0806, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0266, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0285, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0087, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0441, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1964, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1097, 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(0.0229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1682, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0238, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0280, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0360, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0098, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0684, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0495, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0357, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0404, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0184, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0087, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3700, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.46113074204946997, 'recall': 0.5986238532110092, 'f1': 0.5209580838323353, 'number': 436}, 'P': {'precision': 0.6544655929721815, 'recall': 0.599195710455764, 'f1': 0.6256123163051084, 'number': 746}, 'overall_precision': 0.5668534827862289, 'overall_recall': 0.5989847715736041, 'overall_f1': 0.5824763471822295, 'overall_accuracy': 0.8029059829059829}
			------------EPOCH 22---------------
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1444, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0154, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0715, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0148, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0298, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1899, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1127, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0271, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1531, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0228, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0175, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1410, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0208, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0270, 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(0.0338, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0135, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0789, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0274, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0363, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0174, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2860, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.504930966469428, 'recall': 0.5871559633027523, 'f1': 0.542948038176034, 'number': 436}, 'P': {'precision': 0.5992647058823529, 'recall': 0.6554959785522788, 'f1': 0.6261203585147247, 'number': 746}, 'overall_precision': 0.563114134542706, 'overall_recall': 0.6302876480541455, 'overall_f1': 0.5948103792415169, 'overall_accuracy': 0.8248888888888889}
			------------EPOCH 23---------------
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0748, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0437, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0102, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0302, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2507, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1197, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0222, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1690, 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(0.0168, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0966, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0238, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0476, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0516, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0283, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0300, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0204, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2969, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.608, 'recall': 0.5229357798165137, 'f1': 0.562268803945746, 'number': 436}, 'P': {'precision': 0.6056179775280899, 'recall': 0.7225201072386059, 'f1': 0.658924205378973, 'number': 746}, 'overall_precision': 0.6063241106719368, 'overall_recall': 0.6489001692047377, 'overall_f1': 0.6268900694728239, 'overall_accuracy': 0.8337435897435898}
			------------EPOCH 24---------------
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0476, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0211, 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(0.1745, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0815, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0127, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1282, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0142, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1688, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0073, 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(0.0029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0699, 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(0.0325, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0199, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2731, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5090543259557344, 'recall': 0.5802752293577982, 'f1': 0.5423365487674169, 'number': 436}, 'P': {'precision': 0.6454545454545455, 'recall': 0.6662198391420912, 'f1': 0.6556728232189973, 'number': 746}, 'overall_precision': 0.5919494869771112, 'overall_recall': 0.6345177664974619, 'overall_f1': 0.6124948958758676, 'overall_accuracy': 0.8255726495726495}
			------------EPOCH 25---------------
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0681, 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(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0246, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1450, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0671, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0993, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0130, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0950, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0173, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0373, 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(0.0337, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0439, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0141, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0188, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0186, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2115, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5262054507337526, 'recall': 0.5756880733944955, 'f1': 0.5498357064622126, 'number': 436}, 'P': {'precision': 0.6236162361623616, 'recall': 0.6796246648793566, 'f1': 0.6504169339320076, 'number': 746}, 'overall_precision': 0.5875968992248062, 'overall_recall': 0.6412859560067682, 'overall_f1': 0.6132686084142394, 'overall_accuracy': 0.8317606837606838}
			------------EPOCH 26---------------
Loss:  tensor(0.0001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0505, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0247, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0023, 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(0.0174, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1311, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0546, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0922, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0123, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0909, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0211, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0409, 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(0.0284, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0364, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0214, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0164, 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(0.1805, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5339168490153173, 'recall': 0.5596330275229358, 'f1': 0.5464725643896977, 'number': 436}, 'P': {'precision': 0.6225490196078431, 'recall': 0.6809651474530831, 'f1': 0.6504481434058899, 'number': 746}, 'overall_precision': 0.5907305577376276, 'overall_recall': 0.6362098138747885, 'overall_f1': 0.6126272912423625, 'overall_accuracy': 0.8285128205128205}
			------------EPOCH 27---------------
Loss:  tensor(0.0001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0418, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, 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(0.0182, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0125, 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(0.1162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0534, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0856, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1228, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0261, 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(0.0267, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0267, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0160, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0189, 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(0.1631, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5053995680345572, 'recall': 0.536697247706422, 'f1': 0.5205784204671857, 'number': 436}, 'P': {'precision': 0.5951557093425606, 'recall': 0.6916890080428955, 'f1': 0.6398016119032859, 'number': 746}, 'overall_precision': 0.5639097744360902, 'overall_recall': 0.6345177664974619, 'overall_f1': 0.5971337579617834, 'overall_accuracy': 0.8341196581196582}
			------------EPOCH 28---------------
Loss:  tensor(0.0001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0419, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0176, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0139, 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(0.1036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0595, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0795, 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(0.0102, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0899, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0201, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0054, 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(0.0415, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0249, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0181, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0204, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1486, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.543046357615894, 'recall': 0.5642201834862385, 'f1': 0.5534308211473565, 'number': 436}, 'P': {'precision': 0.6271393643031785, 'recall': 0.6876675603217158, 'f1': 0.6560102301790282, 'number': 746}, 'overall_precision': 0.5971675845790716, 'overall_recall': 0.6421319796954315, 'overall_f1': 0.6188340807174888, 'overall_accuracy': 0.835076923076923}
			------------EPOCH 29---------------
Loss:  tensor(8.9789e-05, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0357, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0019, 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(0.0122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0102, 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(0.0872, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0385, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0709, 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(0.0092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1176, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0143, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0276, 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(0.0285, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0261, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0244, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0176, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1391, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5178197064989518, 'recall': 0.5665137614678899, 'f1': 0.5410733844468785, 'number': 436}, 'P': {'precision': 0.6294478527607362, 'recall': 0.6876675603217158, 'f1': 0.6572709801409353, 'number': 746}, 'overall_precision': 0.5882352941176471, 'overall_recall': 0.6429780033840947, 'overall_f1': 0.6143896523848019, 'overall_accuracy': 0.83391452991453}
			------------EPOCH 30---------------
Loss:  tensor(9.3793e-05, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0267, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0104, 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(0.1145, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0512, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0832, 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(0.0085, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0853, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0186, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0052, 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(0.0423, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0231, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0262, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0133, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0169, 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(0.1907, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5403422982885085, 'recall': 0.5068807339449541, 'f1': 0.523076923076923, 'number': 436}, 'P': {'precision': 0.586015538290788, 'recall': 0.707774798927614, 'f1': 0.6411657559198543, 'number': 746}, 'overall_precision': 0.5717557251908397, 'overall_recall': 0.6336717428087987, 'overall_f1': 0.601123595505618, 'overall_accuracy': 0.8305299145299145}


		-------------RUN 5-----------
			------------EPOCH 1---------------
Loss:  tensor(3.5051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4733, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2160, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5528, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9311, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4835, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6601, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1841, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3658, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3391, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2755, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3739, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2338, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1628, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9532, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0878, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0860, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0540, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6754, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4132, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2621, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8645, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6132, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8190, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4385, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7244, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5825, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2852, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7337, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5522, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2205128205128205, 'recall': 0.29723502304147464, 'f1': 0.253189401373896, 'number': 434}, 'P': {'precision': 0.3308457711442786, 'recall': 0.36538461538461536, 'f1': 0.3472584856396867, 'number': 728}, 'overall_precision': 0.2843772498200144, 'overall_recall': 0.3399311531841652, 'overall_f1': 0.30968247745981964, 'overall_accuracy': 0.711907719609583}
			------------EPOCH 2---------------
Loss:  tensor(1.6039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2548, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6701, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5488, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6976, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3224, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3417, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9715, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1145, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6257, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5538, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6436, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2896, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2343, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2370, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2847, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2430, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2848, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2411, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4380, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8959, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1883, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2513, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0851, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7452, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2629, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3300, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1857, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2295, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2645, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2404, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7784, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1443, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4461805555555556, 'recall': 0.5921658986175116, 'f1': 0.508910891089109, 'number': 434}, 'P': {'precision': 0.4782051282051282, 'recall': 0.5123626373626373, 'f1': 0.49469496021220155, 'number': 728}, 'overall_precision': 0.4646017699115044, 'overall_recall': 0.5421686746987951, 'overall_f1': 0.5003971405877681, 'overall_accuracy': 0.8031943212067436}
			------------EPOCH 3---------------
Loss:  tensor(1.0021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1748, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0799, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2685, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1770, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9501, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9732, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7251, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0826, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1279, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9958, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1260, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9496, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2382, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8669, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1765, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8755, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0293, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8490, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8612, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8450, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7566, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0954, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8520, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8207, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9077, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9328, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2652, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9724, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7566, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7891, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5380, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8429, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5588235294117647, 'recall': 0.39400921658986177, 'f1': 0.4621621621621622, 'number': 434}, 'P': {'precision': 0.5383828045035824, 'recall': 0.7225274725274725, 'f1': 0.617008797653959, 'number': 728}, 'overall_precision': 0.5432579890880749, 'overall_recall': 0.5998278829604131, 'overall_f1': 0.5701431492842536, 'overall_accuracy': 0.8259094942324756}
			------------EPOCH 4---------------
Loss:  tensor(0.9815, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1823, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9318, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0697, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8922, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7418, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5737, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0515, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7857, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7717, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9617, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7641, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1673, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6491, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1647, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6381, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8802, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6591, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6449, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2576, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0674, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6769, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6477, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3912, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7517, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2649, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7603, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9540, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2132, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5991, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6591, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3879, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6657, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5422535211267606, 'recall': 0.532258064516129, 'f1': 0.5372093023255815, 'number': 434}, 'P': {'precision': 0.5878995433789954, 'recall': 0.7074175824175825, 'f1': 0.6421446384039899, 'number': 728}, 'overall_precision': 0.5729646697388633, 'overall_recall': 0.6419965576592083, 'overall_f1': 0.6055194805194805, 'overall_accuracy': 0.8439396628216504}
			------------EPOCH 5---------------
Loss:  tensor(0.5991, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1130, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0904, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6998, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5468, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3753, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0340, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4903, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7617, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5292, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1532, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4978, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1385, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5285, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7576, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6495, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4244, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4265, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4607, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1928, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4682, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0509, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4920, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4164, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4858, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3295, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1811, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5948, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6398, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1821, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4695, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2673, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5113, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5102639296187683, 'recall': 0.4009216589861751, 'f1': 0.4490322580645162, 'number': 434}, 'P': {'precision': 0.44577205882352944, 'recall': 0.6662087912087912, 'f1': 0.5341409691629957, 'number': 728}, 'overall_precision': 0.46116165150454863, 'overall_recall': 0.5671256454388984, 'overall_f1': 0.5086839058278657, 'overall_accuracy': 0.8262644188110027}
			------------EPOCH 6---------------
Loss:  tensor(0.5533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1084, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6509, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8963, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4622, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3452, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3913, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2480, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0186, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4240, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4671, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5330, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1298, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3798, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3288, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5462, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5177, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2718, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2468, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2947, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3770, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0369, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3934, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4293, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3904, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1757, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4000, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1367, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3667, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5574, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1644, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3654, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2804, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2261, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4146, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5581947743467933, 'recall': 0.5414746543778802, 'f1': 0.5497076023391813, 'number': 434}, 'P': {'precision': 0.5604026845637584, 'recall': 0.6881868131868132, 'f1': 0.6177558569667078, 'number': 728}, 'overall_precision': 0.5596958174904944, 'overall_recall': 0.6333907056798623, 'overall_f1': 0.5942672587807832, 'overall_accuracy': 0.8291748003549245}
			------------EPOCH 7---------------
Loss:  tensor(0.2594, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0493, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2586, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3264, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0441, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3651, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2247, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2851, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2318, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2396, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2868, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4550, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2731, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2136, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0789, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3397, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7370, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7127, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1929, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1758, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2105, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0830, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2893, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0231, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2932, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2767, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3400, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2683, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1756, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3977, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1529, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3135, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5470, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1617, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2440, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1742, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1826, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3941, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.39752407152682256, 'recall': 0.6658986175115207, 'f1': 0.4978466838931954, 'number': 434}, 'P': {'precision': 0.632688927943761, 'recall': 0.4945054945054945, 'f1': 0.5551272166538166, 'number': 728}, 'overall_precision': 0.5007716049382716, 'overall_recall': 0.5585197934595525, 'overall_f1': 0.5280716029292107, 'overall_accuracy': 0.772564330079858}
			------------EPOCH 8---------------
Loss:  tensor(0.6015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0611, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4727, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4939, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3703, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1981, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2208, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0181, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3809, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7309, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3757, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1296, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0534, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2331, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2684, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2163, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2892, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1228, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3990, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0390, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3340, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2957, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2214, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2300, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1445, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2907, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1879, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3318, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5665, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1622, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2572, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3345, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1825, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3541, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5031712473572939, 'recall': 0.5483870967741935, 'f1': 0.5248070562293276, 'number': 434}, 'P': {'precision': 0.6179361179361179, 'recall': 0.6909340659340659, 'f1': 0.6523994811932555, 'number': 728}, 'overall_precision': 0.5757575757575758, 'overall_recall': 0.6376936316695353, 'overall_f1': 0.6051449571253573, 'overall_accuracy': 0.8375865128660159}
			------------EPOCH 9---------------
Loss:  tensor(0.1984, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0316, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1769, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2280, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5986, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2398, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1625, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1762, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1677, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0107, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1955, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1691, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3649, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2421, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0803, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1503, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0638, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1769, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3957, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3739, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1861, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1342, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1591, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0505, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1545, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0520, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1627, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1149, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1770, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3848, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1236, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0833, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0878, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1494, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5033860045146726, 'recall': 0.5138248847926268, 'f1': 0.5085518814139112, 'number': 434}, 'P': {'precision': 0.6138259833134684, 'recall': 0.7074175824175825, 'f1': 0.6573069559668155, 'number': 728}, 'overall_precision': 0.5756630265210608, 'overall_recall': 0.6351118760757315, 'overall_f1': 0.6039279869067103, 'overall_accuracy': 0.8327950310559006}
			------------EPOCH 10---------------
Loss:  tensor(0.1242, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0264, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1623, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1166, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4547, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2516, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0872, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0886, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0905, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0616, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0634, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2142, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0721, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0660, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1166, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0635, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0809, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1277, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0231, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1927, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0115, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1240, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0843, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1184, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0341, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1236, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3400, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1142, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0357, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0455, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0768, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5536159600997507, 'recall': 0.511520737327189, 'f1': 0.5317365269461078, 'number': 434}, 'P': {'precision': 0.6302816901408451, 'recall': 0.7376373626373627, 'f1': 0.679746835443038, 'number': 728}, 'overall_precision': 0.6057462090981645, 'overall_recall': 0.653184165232358, 'overall_f1': 0.6285714285714286, 'overall_accuracy': 0.8370541259982254}
			------------EPOCH 11---------------
Loss:  tensor(0.0763, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0781, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0636, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4401, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2766, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0504, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0554, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0621, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0634, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0382, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1594, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0374, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0634, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0387, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0147, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2716, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2774, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0337, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0493, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0597, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1842, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0891, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0522, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0419, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0335, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0177, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0403, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0374, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0346, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0818, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.536734693877551, 'recall': 0.6059907834101382, 'f1': 0.5692640692640691, 'number': 434}, 'P': {'precision': 0.6613965744400527, 'recall': 0.6895604395604396, 'f1': 0.6751849361129791, 'number': 728}, 'overall_precision': 0.6124899919935949, 'overall_recall': 0.6583476764199656, 'overall_f1': 0.6345914558274576, 'overall_accuracy': 0.8404259094942325}
			------------EPOCH 12---------------
Loss:  tensor(0.0580, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0364, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3525, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0174, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0246, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0344, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0302, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0208, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1574, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0244, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0331, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0165, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0333, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2367, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2750, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0306, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0518, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0426, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1643, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0679, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0581, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0406, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0143, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0916, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0443, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3380, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0833, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0135, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0160, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0480, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0548, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5492957746478874, 'recall': 0.5391705069124424, 'f1': 0.5441860465116279, 'number': 434}, 'P': {'precision': 0.6267515923566879, 'recall': 0.6758241758241759, 'f1': 0.6503635161929939, 'number': 728}, 'overall_precision': 0.5995045417010735, 'overall_recall': 0.6247848537005164, 'overall_f1': 0.6118836915297092, 'overall_accuracy': 0.8431943212067435}
			------------EPOCH 13---------------
Loss:  tensor(0.0721, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0475, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0676, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3377, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2805, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0201, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0274, 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(0.0271, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0116, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1321, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0332, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0272, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0320, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2355, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2499, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0299, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0445, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0341, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1440, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0661, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0411, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0487, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0308, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0695, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0862, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0367, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2668, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0460, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0207, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0366, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5812182741116751, 'recall': 0.5276497695852534, 'f1': 0.5531400966183576, 'number': 434}, 'P': {'precision': 0.6292682926829268, 'recall': 0.7087912087912088, 'f1': 0.6666666666666666, 'number': 728}, 'overall_precision': 0.613673805601318, 'overall_recall': 0.6411359724612736, 'overall_f1': 0.6271043771043772, 'overall_accuracy': 0.8423779946761313}
			------------EPOCH 14---------------
Loss:  tensor(0.0397, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0213, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0251, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1690, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0182, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0210, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0285, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0518, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1421, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0242, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0350, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0204, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0338, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0218, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0484, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0494, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0337, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0196, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0150, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0796, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0316, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3367, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0461, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0203, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0271, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0325, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0580, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.6022099447513812, 'recall': 0.5023041474654378, 'f1': 0.5477386934673367, 'number': 434}, 'P': {'precision': 0.6478527607361964, 'recall': 0.7252747252747253, 'f1': 0.6843810758263124, 'number': 728}, 'overall_precision': 0.6338147833474936, 'overall_recall': 0.6419965576592083, 'overall_f1': 0.6378794356562633, 'overall_accuracy': 0.8386157941437444}
			------------EPOCH 15---------------
Loss:  tensor(0.0325, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0247, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3290, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1458, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0144, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0188, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0201, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0128, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0361, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0236, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2184, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2361, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0321, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0353, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1481, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0660, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0417, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0280, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0665, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0827, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0297, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2775, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0475, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0165, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0228, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0455, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5160642570281124, 'recall': 0.5921658986175116, 'f1': 0.5515021459227468, 'number': 434}, 'P': {'precision': 0.6511318242343542, 'recall': 0.6717032967032966, 'f1': 0.6612576064908722, 'number': 728}, 'overall_precision': 0.5972778222578062, 'overall_recall': 0.6419965576592083, 'overall_f1': 0.618830360846122, 'overall_accuracy': 0.8307719609582963}
			------------EPOCH 16---------------
Loss:  tensor(0.0365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0241, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0450, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3736, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1885, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0157, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0206, 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(0.0106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0336, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0123, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2587, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2717, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0184, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0242, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1596, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0509, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0369, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0128, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0128, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0512, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0824, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0271, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2507, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0377, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0176, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0224, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0388, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.47113594040968343, 'recall': 0.5829493087557603, 'f1': 0.5211122554067971, 'number': 434}, 'P': {'precision': 0.6005025125628141, 'recall': 0.6565934065934066, 'f1': 0.6272965879265092, 'number': 728}, 'overall_precision': 0.5483870967741935, 'overall_recall': 0.6290877796901894, 'overall_f1': 0.5859719438877755, 'overall_accuracy': 0.8349600709849158}
			------------EPOCH 17---------------
Loss:  tensor(0.0430, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0136, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0222, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4936, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1187, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0126, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0147, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0727, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0199, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2649, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0173, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0143, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1185, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0526, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0247, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0214, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0076, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0661, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0805, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3351, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0713, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0133, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0329, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0453, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5597014925373134, 'recall': 0.5184331797235023, 'f1': 0.5382775119617224, 'number': 434}, 'P': {'precision': 0.6107986501687289, 'recall': 0.7458791208791209, 'f1': 0.6716141001855287, 'number': 728}, 'overall_precision': 0.5948876839659178, 'overall_recall': 0.6609294320137694, 'overall_f1': 0.6261720342437832, 'overall_accuracy': 0.8470984915705413}
			------------EPOCH 18---------------
Loss:  tensor(0.0252, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0084, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2984, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1631, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0107, 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(0.0046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0867, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0395, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2169, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2891, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0112, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0932, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0298, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0169, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0118, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0452, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0541, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2187, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0330, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0331, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5365853658536586, 'recall': 0.5576036866359447, 'f1': 0.5468926553672316, 'number': 434}, 'P': {'precision': 0.6348808030112923, 'recall': 0.695054945054945, 'f1': 0.6636065573770492, 'number': 728}, 'overall_precision': 0.5993589743589743, 'overall_recall': 0.6437177280550774, 'overall_f1': 0.620746887966805, 'overall_accuracy': 0.8370541259982254}
			------------EPOCH 19---------------
Loss:  tensor(0.0260, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2640, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1181, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0715, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0254, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1444, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1971, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0085, 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(0.0936, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0264, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0293, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0443, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0493, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0118, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1708, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0292, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0239, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5064377682403434, 'recall': 0.543778801843318, 'f1': 0.5244444444444444, 'number': 434}, 'P': {'precision': 0.6177514792899408, 'recall': 0.717032967032967, 'f1': 0.663699936427209, 'number': 728}, 'overall_precision': 0.5781845919145691, 'overall_recall': 0.6523235800344234, 'overall_f1': 0.6130206227254348, 'overall_accuracy': 0.8444010647737356}
			------------EPOCH 20---------------
Loss:  tensor(0.0134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2186, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0855, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0568, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0224, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1324, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1594, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0090, 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(0.0662, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0157, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0214, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0298, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0413, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0114, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1739, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0156, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0182, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5361990950226244, 'recall': 0.5460829493087558, 'f1': 0.541095890410959, 'number': 434}, 'P': {'precision': 0.6286057692307693, 'recall': 0.7184065934065934, 'f1': 0.6705128205128206, 'number': 728}, 'overall_precision': 0.5965463108320251, 'overall_recall': 0.6540447504302926, 'overall_f1': 0.6239737274220033, 'overall_accuracy': 0.8443655723158828}
			------------EPOCH 21---------------
Loss:  tensor(0.0195, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0935, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0366, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0195, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1144, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1275, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0853, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0132, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0203, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0084, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0356, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0085, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1350, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0231, 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(0.0067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0098, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5299334811529933, 'recall': 0.5506912442396313, 'f1': 0.5401129943502825, 'number': 434}, 'P': {'precision': 0.6478696741854637, 'recall': 0.7101648351648352, 'f1': 0.6775884665792924, 'number': 728}, 'overall_precision': 0.6052842273819056, 'overall_recall': 0.6506024096385542, 'overall_f1': 0.6271256739941933, 'overall_accuracy': 0.842058562555457}
			------------EPOCH 22---------------
Loss:  tensor(0.0185, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1852, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0356, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0183, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0028, 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(0.1222, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1419, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0670, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0210, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0210, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1479, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0344, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0141, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.502127659574468, 'recall': 0.543778801843318, 'f1': 0.5221238938053097, 'number': 434}, 'P': {'precision': 0.6219806763285024, 'recall': 0.7074175824175825, 'f1': 0.6619537275064268, 'number': 728}, 'overall_precision': 0.5785824345146379, 'overall_recall': 0.6462994836488812, 'overall_f1': 0.6105691056910568, 'overall_accuracy': 0.844223602484472}
			------------EPOCH 23---------------
Loss:  tensor(0.0174, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0986, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0022, 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(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0246, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0236, 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(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1251, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1545, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0039, 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(0.0590, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0129, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0149, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0293, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1199, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0121, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5434782608695652, 'recall': 0.5184331797235023, 'f1': 0.5306603773584906, 'number': 434}, 'P': {'precision': 0.6063084112149533, 'recall': 0.7129120879120879, 'f1': 0.6553030303030304, 'number': 728}, 'overall_precision': 0.5858267716535434, 'overall_recall': 0.6402753872633391, 'overall_f1': 0.611842105263158, 'overall_accuracy': 0.8407453416149069}
			------------EPOCH 24---------------
Loss:  tensor(0.0145, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1938, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0786, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0201, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0914, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0528, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0294, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0174, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0073, 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(0.0308, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0719, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1733, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0107, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5599022004889975, 'recall': 0.5276497695852534, 'f1': 0.5432977461447213, 'number': 434}, 'P': {'precision': 0.6006864988558352, 'recall': 0.7211538461538461, 'f1': 0.6554307116104868, 'number': 728}, 'overall_precision': 0.5876851130163678, 'overall_recall': 0.648881239242685, 'overall_f1': 0.6167689161554192, 'overall_accuracy': 0.8460692102928128}
			------------EPOCH 25---------------
Loss:  tensor(0.0128, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1767, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0805, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0037, 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(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0226, 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(0.0022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1467, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0564, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0142, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0166, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0159, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0336, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1362, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0274, 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(0.0089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0159, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5297872340425532, 'recall': 0.5737327188940092, 'f1': 0.5508849557522124, 'number': 434}, 'P': {'precision': 0.6508344030808729, 'recall': 0.6964285714285714, 'f1': 0.6728599867285998, 'number': 728}, 'overall_precision': 0.6052842273819056, 'overall_recall': 0.6506024096385542, 'overall_f1': 0.6271256739941933, 'overall_accuracy': 0.8422715173025732}
			------------EPOCH 26---------------
Loss:  tensor(0.0197, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2087, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1077, 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(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0246, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0827, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0974, 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(0.0024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0530, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0214, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0275, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0502, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1381, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0344, 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>)
Loss:  tensor(0.0086, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5344036697247706, 'recall': 0.5368663594470046, 'f1': 0.5356321839080459, 'number': 434}, 'P': {'precision': 0.62590799031477, 'recall': 0.7101648351648352, 'f1': 0.6653796653796653, 'number': 728}, 'overall_precision': 0.5942947702060222, 'overall_recall': 0.6454388984509466, 'overall_f1': 0.6188118811881189, 'overall_accuracy': 0.8437622005323868}
			------------EPOCH 27---------------
Loss:  tensor(0.0158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1715, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0786, 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(0.0016, 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(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0321, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0313, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1286, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1338, 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(0.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0625, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0155, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0118, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0322, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0976, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0106, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.565, 'recall': 0.5207373271889401, 'f1': 0.5419664268585132, 'number': 434}, 'P': {'precision': 0.6158323632130385, 'recall': 0.7266483516483516, 'f1': 0.6666666666666666, 'number': 728}, 'overall_precision': 0.5996822875297856, 'overall_recall': 0.6497418244406197, 'overall_f1': 0.6237092110698059, 'overall_accuracy': 0.8437267080745342}
			------------EPOCH 28---------------
Loss:  tensor(0.0162, 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(0.0035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1417, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0810, 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(0.0001, 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(0.0261, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0187, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0865, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0858, 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(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0358, 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(0.0149, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0115, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0308, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0951, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0062, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5010660980810234, 'recall': 0.5414746543778802, 'f1': 0.5204872646733112, 'number': 434}, 'P': {'precision': 0.6295843520782396, 'recall': 0.7074175824175825, 'f1': 0.6662354463130661, 'number': 728}, 'overall_precision': 0.5827505827505828, 'overall_recall': 0.6454388984509466, 'overall_f1': 0.6124948958758677, 'overall_accuracy': 0.843620230700976}
			------------EPOCH 29---------------
Loss:  tensor(0.0165, 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(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0913, 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(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0155, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0092, 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(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0585, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0771, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0292, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0118, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0077, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0236, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0685, 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(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0029, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5245535714285714, 'recall': 0.5414746543778802, 'f1': 0.5328798185941044, 'number': 434}, 'P': {'precision': 0.6143867924528302, 'recall': 0.7156593406593407, 'f1': 0.6611675126903553, 'number': 728}, 'overall_precision': 0.5833333333333334, 'overall_recall': 0.6506024096385542, 'overall_f1': 0.6151342554922701, 'overall_accuracy': 0.8464951197870453}
			------------EPOCH 30---------------
Loss:  tensor(0.0131, 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(0.1068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1019, 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(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.6094e-05, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, 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(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0107, 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(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0695, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0246, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0136, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0622, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0144, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0029, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5324074074074074, 'recall': 0.5299539170506913, 'f1': 0.5311778290993072, 'number': 434}, 'P': {'precision': 0.6239419588875453, 'recall': 0.7087912087912088, 'f1': 0.6636655948553054, 'number': 728}, 'overall_precision': 0.5925337569499602, 'overall_recall': 0.6419965576592083, 'overall_f1': 0.6162742668318877, 'overall_accuracy': 0.8409937888198757}
	Train size: 50 Test size: 50


		-------------RUN 1-----------
			------------EPOCH 1---------------
Loss:  tensor(0.2375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4202, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7553, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9513, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2404, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9682, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6432, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7959, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7102, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5289, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5505, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5547, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2828, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3258, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3938, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2866, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5234, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2333, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1532, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8912, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0981, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.19047619047619047, 'recall': 0.003389830508474576, 'f1': 0.006661115736885928, 'number': 1180}, 'P': {'precision': 0.4087013843111404, 'recall': 0.6129510627780524, 'f1': 0.4904093335969943, 'number': 2023}, 'overall_precision': 0.4072013093289689, 'overall_recall': 0.3883858882297846, 'overall_f1': 0.39757110898050496, 'overall_accuracy': 0.6926239814907957}
			------------EPOCH 2---------------
Loss:  tensor(0.1559, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9876, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4807, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8789, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1242, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6453, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7579, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9520, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6821, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7766, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5412, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2991, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5341, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8171, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3926, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1286, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5392, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5544, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6674, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4659, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3582887700534759, 'recall': 0.1135593220338983, 'f1': 0.17245817245817247, 'number': 1180}, 'P': {'precision': 0.37370447142434116, 'recall': 0.6238260009886307, 'f1': 0.46740740740740744, 'number': 2023}, 'overall_precision': 0.37216742202079445, 'overall_recall': 0.43584139868872934, 'overall_f1': 0.4014955421340236, 'overall_accuracy': 0.7278820038225531}
			------------EPOCH 3---------------
Loss:  tensor(0.1337, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4300, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3997, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4750, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0813, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4739, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2931, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4299, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2548, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3170, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2531, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1247, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2280, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1575, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2908, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1295, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2187, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2788, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2786, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8233, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1341, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4800, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0302, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.35443037974683544, 'recall': 0.45084745762711864, 'f1': 0.3968668407310705, 'number': 1180}, 'P': {'precision': 0.5219490781387182, 'recall': 0.587740978744439, 'f1': 0.552894675656824, 'number': 2023}, 'overall_precision': 0.4554114845197142, 'overall_recall': 0.5373087730252888, 'overall_f1': 0.4929819535949585, 'overall_accuracy': 0.786653757167287}
			------------EPOCH 4---------------
Loss:  tensor(0.0885, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0156, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3280, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1341, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0551, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3412, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0648, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1500, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9550, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0252, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8813, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9516, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0398, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8812, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2193, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8686, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1799, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9381, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6303, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8105, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8132, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3336, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8069, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.36156696664902066, 'recall': 0.5788135593220339, 'f1': 0.4450961225154774, 'number': 1180}, 'P': {'precision': 0.5890949971894323, 'recall': 0.5180425111220959, 'f1': 0.5512887953708574, 'number': 2023}, 'overall_precision': 0.4719193020719738, 'overall_recall': 0.5404308460817984, 'overall_f1': 0.5038567894047445, 'overall_accuracy': 0.7820264560909366}
			------------EPOCH 5---------------
Loss:  tensor(0.0665, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8458, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2722, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9090, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0400, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2623, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7129, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7592, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6637, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6997, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6564, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5994, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6957, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8565, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7374, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1702, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6806, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1301, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7887, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1700, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4174, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5195, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5498, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2428, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5345, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5459905660377359, 'recall': 0.3923728813559322, 'f1': 0.45660749506903353, 'number': 1180}, 'P': {'precision': 0.5585271317829458, 'recall': 0.7123084527928819, 'f1': 0.6261134043015425, 'number': 2023}, 'overall_precision': 0.5554259043173863, 'overall_recall': 0.5944427099594131, 'overall_f1': 0.5742723571105415, 'overall_accuracy': 0.8029750528115883}
			------------EPOCH 6---------------
Loss:  tensor(0.0506, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7298, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2601, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6300, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0397, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1810, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7149, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5857, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5132, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5712, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4768, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3490, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5814, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8590, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1522, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4747, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5234, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1595, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2929, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3426, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3868, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1574, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4117, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.37071470075537477, 'recall': 0.5406779661016949, 'f1': 0.4398483281627025, 'number': 1180}, 'P': {'precision': 0.6188459502382213, 'recall': 0.5778546712802768, 'f1': 0.5976482617586911, 'number': 2023}, 'overall_precision': 0.5005540166204986, 'overall_recall': 0.5641586013112707, 'overall_f1': 0.5304564802583296, 'overall_accuracy': 0.7894075042752238}
			------------EPOCH 7---------------
Loss:  tensor(0.0194, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3921, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2346, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0310, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3955, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3750, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2683, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3154, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2895, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2991, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2479, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0832, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2487, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0788, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3516, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1102, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2164, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0921, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2337, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.43071672354948803, 'recall': 0.5347457627118644, 'f1': 0.47712665406427224, 'number': 1180}, 'P': {'precision': 0.6180758017492711, 'recall': 0.6287691547207118, 'f1': 0.6233766233766234, 'number': 2023}, 'overall_precision': 0.5401646324155549, 'overall_recall': 0.5941305026537621, 'overall_f1': 0.5658638120725543, 'overall_accuracy': 0.80336485263052}
			------------EPOCH 8---------------
Loss:  tensor(0.0117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2317, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1632, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3196, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0746, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3159, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1989, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1516, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1585, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3715, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0831, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1576, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0418, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3248, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2181, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2632, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0868, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2326, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.36785069235400364, 'recall': 0.5177966101694915, 'f1': 0.43013023583245336, 'number': 1180}, 'P': {'precision': 0.530782761653474, 'recall': 0.5966386554621849, 'f1': 0.5617872934605539, 'number': 2023}, 'overall_precision': 0.46200762388818295, 'overall_recall': 0.5675928816734311, 'overall_f1': 0.5093863827402633, 'overall_accuracy': 0.7954808369379338}
			------------EPOCH 9---------------
Loss:  tensor(0.0100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1788, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1816, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0503, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1612, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1597, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1755, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1259, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1676, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1390, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1273, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7520, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1763, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1644, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0853, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2697, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1269, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1822, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0470, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1243, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4067398119122257, 'recall': 0.4398305084745763, 'f1': 0.42263843648208466, 'number': 1180}, 'P': {'precision': 0.470523617722446, 'recall': 0.6351952545724172, 'f1': 0.5405973916701725, 'number': 2023}, 'overall_precision': 0.4502121287746444, 'overall_recall': 0.5632219793943178, 'overall_f1': 0.5004160887656034, 'overall_accuracy': 0.8049114777185394}
			------------EPOCH 10---------------
Loss:  tensor(0.0084, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1516, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2829, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0228, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0554, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2212, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1796, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1954, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1728, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4758, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0729, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0634, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0749, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0196, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2135, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0959, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1111, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1476, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0551, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1015, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4568121104185218, 'recall': 0.4347457627118644, 'f1': 0.4455058619192358, 'number': 1180}, 'P': {'precision': 0.5806185567010309, 'recall': 0.6959960454770143, 'f1': 0.6330935251798562, 'number': 2023}, 'overall_precision': 0.5414317925591883, 'overall_recall': 0.5997502341554792, 'overall_f1': 0.5691008739446007, 'overall_accuracy': 0.810871642691882}
			------------EPOCH 11---------------
Loss:  tensor(0.0065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0971, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1275, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1976, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0484, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1247, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0983, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0543, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3605, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0589, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0492, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1346, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0681, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0439, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0309, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0814, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0224, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0919, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5392491467576792, 'recall': 0.4016949152542373, 'f1': 0.46041767848470133, 'number': 1180}, 'P': {'precision': 0.5851765618936748, 'recall': 0.7454275827978251, 'f1': 0.6556521739130435, 'number': 2023}, 'overall_precision': 0.5734953703703703, 'overall_recall': 0.6187948798001873, 'overall_f1': 0.5952845772638534, 'overall_accuracy': 0.803125943064078}
			------------EPOCH 12---------------
Loss:  tensor(0.0100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1376, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0273, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0503, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0452, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0779, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0452, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0544, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0432, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0412, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2513, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0800, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0538, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0936, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0265, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1308, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0619, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0800, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0397, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0322, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5309126594700687, 'recall': 0.45847457627118643, 'f1': 0.49204183719872663, 'number': 1180}, 'P': {'precision': 0.5690327613104524, 'recall': 0.7212061295106278, 'f1': 0.6361456289513844, 'number': 2023}, 'overall_precision': 0.5581914596706671, 'overall_recall': 0.6244146113019045, 'overall_f1': 0.5894488653109342, 'overall_accuracy': 0.8126823257217584}
			------------EPOCH 13---------------
Loss:  tensor(0.0021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0547, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0982, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0515, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0539, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0704, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0771, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0550, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0352, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2441, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0168, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0437, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0275, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0998, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0439, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0549, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0423, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1302, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0127, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1030, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.41961067051189616, 'recall': 0.49322033898305084, 'f1': 0.4534476042072458, 'number': 1180}, 'P': {'precision': 0.5598021582733813, 'recall': 0.6154226396440929, 'f1': 0.5862962090887686, 'number': 2023}, 'overall_precision': 0.5059540293547494, 'overall_recall': 0.5704027474242898, 'overall_f1': 0.5362488993249193, 'overall_accuracy': 0.8012649632833719}
			------------EPOCH 14---------------
Loss:  tensor(0.0043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0855, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0387, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0241, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0359, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0257, 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(0.0373, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2098, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0290, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0435, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0423, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0923, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0434, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0129, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0591, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0202, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4544138929088278, 'recall': 0.5322033898305085, 'f1': 0.4902419984387198, 'number': 1180}, 'P': {'precision': 0.586890901250539, 'recall': 0.6727632229362334, 'f1': 0.6269000460617228, 'number': 2023}, 'overall_precision': 0.5374223182923534, 'overall_recall': 0.620980330939744, 'overall_f1': 0.5761877172653533, 'overall_accuracy': 0.8076777990141837}
			------------EPOCH 15---------------
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0889, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0798, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0248, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0233, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0236, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0169, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1951, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0450, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0187, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0790, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0369, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0164, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0460, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0120, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4823717948717949, 'recall': 0.5101694915254237, 'f1': 0.4958813838550247, 'number': 1180}, 'P': {'precision': 0.6005165733964701, 'recall': 0.6895699456253089, 'f1': 0.6419696272434423, 'number': 2023}, 'overall_precision': 0.5592271072528704, 'overall_recall': 0.6234779893849516, 'overall_f1': 0.589607322113965, 'overall_accuracy': 0.8123931194044864}
			------------EPOCH 16---------------
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0791, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0607, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0139, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0149, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0139, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0111, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2481, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0617, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0184, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0687, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0333, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0379, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0100, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5146612740141557, 'recall': 0.43135593220338986, 'f1': 0.4693407100046105, 'number': 1180}, 'P': {'precision': 0.5568400770712909, 'recall': 0.7142857142857143, 'f1': 0.6258120398440883, 'number': 2023}, 'overall_precision': 0.5452008928571429, 'overall_recall': 0.6100530752419606, 'overall_f1': 0.5758066892588772, 'overall_accuracy': 0.8119907453978473}
			------------EPOCH 17---------------
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0345, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0891, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0665, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0194, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0111, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1369, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0366, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0160, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0494, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0290, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0313, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0085, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.45170454545454547, 'recall': 0.5389830508474577, 'f1': 0.49149922720247297, 'number': 1180}, 'P': {'precision': 0.6109601449275363, 'recall': 0.6668314384577361, 'f1': 0.6376743086740724, 'number': 2023}, 'overall_precision': 0.5489491150442478, 'overall_recall': 0.6197315017171402, 'overall_f1': 0.5821968030503007, 'overall_accuracy': 0.809941152801529}
			------------EPOCH 18---------------
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0311, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1128, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0570, 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(0.0069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0076, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1127, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0345, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0208, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0479, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0267, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0224, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0071, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.519893899204244, 'recall': 0.49830508474576274, 'f1': 0.508870618779749, 'number': 1180}, 'P': {'precision': 0.5908150599917252, 'recall': 0.7058823529411765, 'f1': 0.6432432432432433, 'number': 2023}, 'overall_precision': 0.5682074408117249, 'overall_recall': 0.6294099281923197, 'overall_f1': 0.5972448526144275, 'overall_accuracy': 0.8161025047781914}
			------------EPOCH 19---------------
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0285, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0632, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0323, 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(0.0037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0136, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0983, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0319, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0339, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0230, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0063, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5059322033898305, 'recall': 0.5059322033898305, 'f1': 0.5059322033898305, 'number': 1180}, 'P': {'precision': 0.5957088767353808, 'recall': 0.6999505684626792, 'f1': 0.6436363636363638, 'number': 2023}, 'overall_precision': 0.5659263424233905, 'overall_recall': 0.6284733062753668, 'overall_f1': 0.5955621301775148, 'overall_accuracy': 0.8155869630821849}
			------------EPOCH 20---------------
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0262, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0546, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0275, 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(0.0028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0081, 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(0.0071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0866, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0301, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0180, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0288, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0057, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.49957228400342174, 'recall': 0.49491525423728816, 'f1': 0.49723286504895703, 'number': 1180}, 'P': {'precision': 0.5816831683168316, 'recall': 0.6969846762234305, 'f1': 0.6341353721610073, 'number': 2023}, 'overall_precision': 0.5549679933203451, 'overall_recall': 0.6225413674679987, 'overall_f1': 0.5868157739846969, 'overall_accuracy': 0.8146564731918318}
			------------EPOCH 21---------------
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0224, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0527, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0239, 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(0.0024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0308, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0174, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0288, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0201, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0053, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.516245487364621, 'recall': 0.4847457627118644, 'f1': 0.5000000000000001, 'number': 1180}, 'P': {'precision': 0.5875566543057272, 'recall': 0.7048937221947602, 'f1': 0.6408988764044944, 'number': 2023}, 'overall_precision': 0.5652050919377652, 'overall_recall': 0.6237901966906025, 'overall_f1': 0.5930543187889582, 'overall_accuracy': 0.8144427120008048}
			------------EPOCH 22---------------
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0181, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0458, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0049, 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(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0955, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0188, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0231, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0234, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0194, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0051, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4471488178025035, 'recall': 0.5449152542372881, 'f1': 0.4912146676852559, 'number': 1180}, 'P': {'precision': 0.610036832412523, 'recall': 0.6549678695007415, 'f1': 0.6317044100119189, 'number': 2023}, 'overall_precision': 0.5451523545706372, 'overall_recall': 0.614423977521074, 'overall_f1': 0.5777190664905328, 'overall_accuracy': 0.8098657076752842}
			------------EPOCH 23---------------
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1180, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0484, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0021, 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(0.0059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0897, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0266, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0177, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0370, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0249, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0027, 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(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0049, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5317307692307692, 'recall': 0.46864406779661016, 'f1': 0.4981981981981982, 'number': 1180}, 'P': {'precision': 0.5726837060702875, 'recall': 0.7088482451804251, 'f1': 0.633532140490391, 'number': 2023}, 'overall_precision': 0.5606659142212189, 'overall_recall': 0.6203559163284421, 'overall_f1': 0.5890025196383578, 'overall_accuracy': 0.8120661905240921}
			------------EPOCH 24---------------
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0830, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0267, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0906, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0295, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0290, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0183, 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>)
				 {'C': {'precision': 0.4743276283618582, 'recall': 0.49322033898305084, 'f1': 0.4835895305359369, 'number': 1180}, 'P': {'precision': 0.5886373344724477, 'recall': 0.6811665842807711, 'f1': 0.6315307057745188, 'number': 2023}, 'overall_precision': 0.5493273542600897, 'overall_recall': 0.6119263190758664, 'overall_f1': 0.5789395953330378, 'overall_accuracy': 0.8129212352882004}
			------------EPOCH 25---------------
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0506, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0190, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0618, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0279, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0173, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0180, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0203, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0157, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0041, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4873054873054873, 'recall': 0.5042372881355932, 'f1': 0.49562682215743437, 'number': 1180}, 'P': {'precision': 0.59229466553768, 'recall': 0.6915472071181413, 'f1': 0.6380843785632839, 'number': 2023}, 'overall_precision': 0.556516885291655, 'overall_recall': 0.6225413674679987, 'overall_f1': 0.5876805187150015, 'overall_accuracy': 0.8120410421486772}
			------------EPOCH 26---------------
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0370, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0001, 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(0.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0490, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0246, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0157, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0165, 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(0.0162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0033, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5074758135444152, 'recall': 0.48898305084745763, 'f1': 0.49805783340526544, 'number': 1180}, 'P': {'precision': 0.5827486586875774, 'recall': 0.6979733069698467, 'f1': 0.6351776878092669, 'number': 2023}, 'overall_precision': 0.5587078651685393, 'overall_recall': 0.620980330939744, 'overall_f1': 0.5882005027354724, 'overall_accuracy': 0.8128457901619556}
			------------EPOCH 27---------------
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0320, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0133, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, 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(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0356, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0199, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0157, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0159, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0028, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4727838258164852, 'recall': 0.5152542372881356, 'f1': 0.49310624493106237, 'number': 1180}, 'P': {'precision': 0.5979292493528904, 'recall': 0.6851211072664359, 'f1': 0.6385625431928127, 'number': 2023}, 'overall_precision': 0.553274139844617, 'overall_recall': 0.6225413674679987, 'overall_f1': 0.5858674893491993, 'overall_accuracy': 0.8115380746403782}
			------------EPOCH 28---------------
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0303, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0312, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0185, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0152, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0135, 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(0.0164, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0019, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4987012987012987, 'recall': 0.488135593220339, 'f1': 0.49336188436830836, 'number': 1180}, 'P': {'precision': 0.5808580858085809, 'recall': 0.6959960454770143, 'f1': 0.633235889363616, 'number': 2023}, 'overall_precision': 0.5543447890472198, 'overall_recall': 0.6194192944114892, 'overall_f1': 0.5850781480389265, 'overall_accuracy': 0.8135625188612816}
			------------EPOCH 29---------------
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0309, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0118, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.4985e-05, 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(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, 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(0.0030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0279, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0166, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0157, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0159, 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(0.0152, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4704944178628389, 'recall': 0.5, 'f1': 0.4847986852917009, 'number': 1180}, 'P': {'precision': 0.5806178586542531, 'recall': 0.6782006920415224, 'f1': 0.6256269949840402, 'number': 2023}, 'overall_precision': 0.5424384849322643, 'overall_recall': 0.6125507336871683, 'overall_f1': 0.575366568914956, 'overall_accuracy': 0.8113117392616437}
			------------EPOCH 30---------------
Loss:  tensor(0.0001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0282, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0259e-05, 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(0.0012, 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(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0391, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0201, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0169, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0097, 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(0.0156, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4614797864225782, 'recall': 0.5127118644067796, 'f1': 0.4857486953030911, 'number': 1180}, 'P': {'precision': 0.5949587136027814, 'recall': 0.6767177459218982, 'f1': 0.6332099907493062, 'number': 2023}, 'overall_precision': 0.5465116279069767, 'overall_recall': 0.6162972213549797, 'overall_f1': 0.5793103448275863, 'overall_accuracy': 0.8110476813197868}


		-------------RUN 2-----------
			------------EPOCH 1---------------
Loss:  tensor(3.6490, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5111, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2204, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1618, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4376, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7795, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9921, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6274, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5248, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1973, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4207, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5233, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5412, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3847, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1862, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3295, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5895, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6121, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1869, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2331, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1699, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.22833723653395785, 'recall': 0.35135135135135137, 'f1': 0.276792051100071, 'number': 1110}, 'P': {'precision': 0.48833819241982507, 'recall': 0.3434136340338288, 'f1': 0.403250075233223, 'number': 1951}, 'overall_precision': 0.34415584415584416, 'overall_recall': 0.3462920614178373, 'overall_f1': 0.34522064810291486, 'overall_accuracy': 0.6565406821340747}
			------------EPOCH 2---------------
Loss:  tensor(2.0279, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9864, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1418, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9873, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7772, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6272, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3176, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6436, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0945, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6984, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4944, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6854, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2329, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5366, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6944, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2338, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1196, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1295, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5728, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5414, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6169, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.29101194217473286, 'recall': 0.41711711711711713, 'f1': 0.34283598667160314, 'number': 1110}, 'P': {'precision': 0.3889396709323583, 'recall': 0.4361865709892363, 'f1': 0.4112104373036966, 'number': 1951}, 'overall_precision': 0.3477110346652554, 'overall_recall': 0.4292714799085266, 'overall_f1': 0.3842105263157895, 'overall_accuracy': 0.7347107879824655}
			------------EPOCH 3---------------
Loss:  tensor(1.3773, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4696, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1913, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9310, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1596, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3310, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0704, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1671, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1418, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1914, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0574, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3516, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1768, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1123, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1873, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8863, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1421, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1770, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2423, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.33644400785854617, 'recall': 0.6171171171171171, 'f1': 0.4354736172917991, 'number': 1110}, 'P': {'precision': 0.4703511801957398, 'recall': 0.4187596104561763, 'f1': 0.44305856832971796, 'number': 1951}, 'overall_precision': 0.3980917042141532, 'overall_recall': 0.49068931721659587, 'overall_f1': 0.4395668715247293, 'overall_accuracy': 0.7409654656260024}
			------------EPOCH 4---------------
Loss:  tensor(1.0119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2194, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0646, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9538, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0622, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7266, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8537, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0333, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0540, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8886, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8836, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7912, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3934, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1645, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6975, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6943, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8615, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9442, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3436765397273155, 'recall': 0.6585585585585586, 'f1': 0.451652764905777, 'number': 1110}, 'P': {'precision': 0.5359801488833746, 'recall': 0.44284982060481803, 'f1': 0.4849845635700253, 'number': 1951}, 'overall_precision': 0.42658464830168497, 'overall_recall': 0.5210715452466514, 'overall_f1': 0.46911764705882353, 'overall_accuracy': 0.7475676253608468}
			------------EPOCH 5---------------
Loss:  tensor(0.7462, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9565, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0592, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7328, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8403, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6385, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7657, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5432, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5944, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8233, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0344, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6252, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5514, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0506, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0680, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5247, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2682, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1218, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5744, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6496, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6631, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6942, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3744703389830508, 'recall': 0.636936936936937, 'f1': 0.4716477651767845, 'number': 1110}, 'P': {'precision': 0.5988629185091598, 'recall': 0.4859046642747309, 'f1': 0.5365025466893039, 'number': 1951}, 'overall_precision': 0.47680783635839813, 'overall_recall': 0.5406729826853969, 'overall_f1': 0.5067360685854256, 'overall_accuracy': 0.7646610713140168}
			------------EPOCH 6---------------
Loss:  tensor(0.5654, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7646, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0446, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6572, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7939, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5944, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4914, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3364, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8202, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0195, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5169, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6168, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6123, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8577, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0629, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2753, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0858, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4518, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5778, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4477, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5103, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4404494382022472, 'recall': 0.5297297297297298, 'f1': 0.4809815950920246, 'number': 1110}, 'P': {'precision': 0.627254509018036, 'recall': 0.6417221937467965, 'f1': 0.6344058778819357, 'number': 1951}, 'overall_precision': 0.5523866706694687, 'overall_recall': 0.6011107481215289, 'overall_f1': 0.5757196495619524, 'overall_accuracy': 0.8153800919491072}
			------------EPOCH 7---------------
Loss:  tensor(0.5087, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5442, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0372, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4807, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3645, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3853, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2333, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5443, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0160, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2900, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3332, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3434, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2800, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6735, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0476, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2896, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1458, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0627, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3587, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2581, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2631, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3844, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4301639344262295, 'recall': 0.590990990990991, 'f1': 0.4979127134724857, 'number': 1110}, 'P': {'precision': 0.5994050570153694, 'recall': 0.619682214249103, 'f1': 0.609375, 'number': 1951}, 'overall_precision': 0.5265386787125917, 'overall_recall': 0.6092780137210062, 'overall_f1': 0.5648947448129639, 'overall_accuracy': 0.8075751095905057}
			------------EPOCH 8---------------
Loss:  tensor(0.2639, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3561, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0171, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4791, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2595, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1562, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2126, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3877, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0148, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1979, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1924, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1862, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0231, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1648, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0891, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0282, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3356, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1798, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1949, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3222, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3586659376708584, 'recall': 0.590990990990991, 'f1': 0.4464103436543042, 'number': 1110}, 'P': {'precision': 0.5586374695863747, 'recall': 0.5884161968221425, 'f1': 0.5731402895656516, 'number': 1951}, 'overall_precision': 0.46446961894953653, 'overall_recall': 0.5893498856582816, 'overall_f1': 0.5195104391648669, 'overall_accuracy': 0.8049957232973377}
			------------EPOCH 9---------------
Loss:  tensor(0.1987, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3140, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0118, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1450, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4137, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0973, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1318, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0927, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1182, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1118, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1477, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5350, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0166, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1840, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0778, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0246, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3228, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2308, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3744997141223556, 'recall': 0.5900900900900901, 'f1': 0.45820216859041624, 'number': 1110}, 'P': {'precision': 0.5414171656686627, 'recall': 0.5561250640697079, 'f1': 0.5486725663716815, 'number': 1951}, 'overall_precision': 0.4636290967226219, 'overall_recall': 0.5684416857236197, 'overall_f1': 0.5107132374523041, 'overall_accuracy': 0.8032449481449803}
			------------EPOCH 10---------------
Loss:  tensor(0.0996, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1691, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0085, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1207, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3991, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2294, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1737, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3275, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0644, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0585, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0825, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0844, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4886, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0227, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0963, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0741, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3761, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2512, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2343, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1921, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2543, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4116022099447514, 'recall': 0.5369369369369369, 'f1': 0.4659890539483972, 'number': 1110}, 'P': {'precision': 0.6310326933056565, 'recall': 0.6232701178882625, 'f1': 0.627127385250129, 'number': 1951}, 'overall_precision': 0.5368888888888889, 'overall_recall': 0.5919634106501144, 'overall_f1': 0.5630826600372902, 'overall_accuracy': 0.8087912969100823}
			------------EPOCH 11---------------
Loss:  tensor(0.0522, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1532, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1248, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0923, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1585, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4426, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2358, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1806, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1845, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1279, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5788, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0085, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0956, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0449, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0114, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1413, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4121, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2386, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3648, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4110854503464203, 'recall': 0.3207207207207207, 'f1': 0.36032388663967607, 'number': 1110}, 'P': {'precision': 0.515527950310559, 'recall': 0.6381342901076371, 'f1': 0.570316078790655, 'number': 1951}, 'overall_precision': 0.48796098750380984, 'overall_recall': 0.523031688990526, 'overall_f1': 0.5048880479344056, 'overall_accuracy': 0.7851090559178873}
			------------EPOCH 12---------------
Loss:  tensor(0.1503, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3568, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1143, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1427, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1240, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2323, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4254, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3691, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2416, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2827, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1918, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0127, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1334, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0635, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2645, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1260, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1139, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2808, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.43660418963616315, 'recall': 0.3567567567567568, 'f1': 0.3926623698562221, 'number': 1110}, 'P': {'precision': 0.5023976392475101, 'recall': 0.6981035366478728, 'f1': 0.5842985842985843, 'number': 1951}, 'overall_precision': 0.4859038142620232, 'overall_recall': 0.5743221169552434, 'overall_f1': 0.5264261116933673, 'overall_accuracy': 0.8107692718913717}
			------------EPOCH 13---------------
Loss:  tensor(0.2337, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3758, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0220, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3369, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5345, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2922, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1608, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3239, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0821, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0642, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0710, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0671, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1324, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0485, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4711, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3451, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4115, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.43027098479841375, 'recall': 0.5864864864864865, 'f1': 0.49637819290888296, 'number': 1110}, 'P': {'precision': 0.6059405940594059, 'recall': 0.6273705791901589, 'f1': 0.6164694031730042, 'number': 1951}, 'overall_precision': 0.5307104443815455, 'overall_recall': 0.6125449199607971, 'overall_f1': 0.5686988171064605, 'overall_accuracy': 0.8124799529562707}
			------------EPOCH 14---------------
Loss:  tensor(0.1158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2315, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0711, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3426, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0566, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0878, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0607, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2984, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1752, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1421, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1483, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1617, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4635, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0791, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0171, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2098, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0651, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0639, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0415, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1872, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5150884495317378, 'recall': 0.44594594594594594, 'f1': 0.4780299372283921, 'number': 1110}, 'P': {'precision': 0.5976451481932602, 'recall': 0.7544848795489493, 'f1': 0.6669687358405074, 'number': 1951}, 'overall_precision': 0.5744742990654206, 'overall_recall': 0.6426004573668735, 'overall_f1': 0.6066306861989206, 'overall_accuracy': 0.8163289853522934}
			------------EPOCH 15---------------
Loss:  tensor(0.1063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1903, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1184, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4515, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0960, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0398, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1978, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0199, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0234, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0293, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0652, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3983, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0546, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2581, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1242, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0615, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2877, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4355179704016913, 'recall': 0.5567567567567567, 'f1': 0.4887307236061685, 'number': 1110}, 'P': {'precision': 0.5987595419847328, 'recall': 0.6432598667350077, 'f1': 0.6202125030887077, 'number': 1951}, 'overall_precision': 0.5328591749644381, 'overall_recall': 0.611891538712839, 'overall_f1': 0.5696472019464721, 'overall_accuracy': 0.8186410777290709}
			------------EPOCH 16---------------
Loss:  tensor(0.0403, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0790, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0149, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2779, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0174, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0308, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1842, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0280, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0218, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3694, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0254, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0184, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1818, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0696, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0335, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0185, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1372, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.44867549668874174, 'recall': 0.4882882882882883, 'f1': 0.46764452113891286, 'number': 1110}, 'P': {'precision': 0.6000848176420696, 'recall': 0.7252690927729369, 'f1': 0.6567649106521234, 'number': 1951}, 'overall_precision': 0.5487941671340437, 'overall_recall': 0.6393335511270827, 'overall_f1': 0.5906141542175948, 'overall_accuracy': 0.8223430984710788}
			------------EPOCH 17---------------
Loss:  tensor(0.0158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2532, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0133, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0251, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0165, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0130, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1505, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0105, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0152, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3116, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0141, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1624, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0384, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0231, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1217, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.48414023372287146, 'recall': 0.5225225225225225, 'f1': 0.5025996533795494, 'number': 1110}, 'P': {'precision': 0.6265765765765766, 'recall': 0.7129677088672476, 'f1': 0.6669863342124192, 'number': 1951}, 'overall_precision': 0.5766530134581627, 'overall_recall': 0.64390721986279, 'overall_f1': 0.6084272264238308, 'overall_accuracy': 0.8276221533197904}
			------------EPOCH 18---------------
Loss:  tensor(0.0062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2257, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0129, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0076, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1255, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2806, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1540, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0473, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0191, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1422, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.456968773800457, 'recall': 0.5405405405405406, 'f1': 0.49525381758151055, 'number': 1110}, 'P': {'precision': 0.6261031119368323, 'recall': 0.690927729369554, 'f1': 0.6569200779727096, 'number': 1951}, 'overall_precision': 0.5620311598384304, 'overall_recall': 0.6363933355112709, 'overall_f1': 0.5969051631683774, 'overall_accuracy': 0.8275686945365124}
			------------EPOCH 19---------------
Loss:  tensor(0.0052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0333, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1092, 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(0.0049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0090, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2538, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0090, 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(0.1381, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0598, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0156, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1505, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4882017900732303, 'recall': 0.5405405405405406, 'f1': 0.5130397605814451, 'number': 1110}, 'P': {'precision': 0.6305644791188618, 'recall': 0.7042542286007176, 'f1': 0.6653753026634383, 'number': 1951}, 'overall_precision': 0.579225352112676, 'overall_recall': 0.6448872917347273, 'overall_f1': 0.6102952542896893, 'overall_accuracy': 0.8296001283010799}
			------------EPOCH 20---------------
Loss:  tensor(0.0039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0308, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1990, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0970, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2413, 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(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1335, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0587, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0126, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1272, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4616576297443842, 'recall': 0.5369369369369369, 'f1': 0.4964598084131611, 'number': 1110}, 'P': {'precision': 0.6301305970149254, 'recall': 0.6924654023577652, 'f1': 0.6598290598290598, 'number': 1951}, 'overall_precision': 0.5668122270742358, 'overall_recall': 0.6360666448872917, 'overall_f1': 0.5994458128078818, 'overall_accuracy': 0.827408318186678}
			------------EPOCH 21---------------
Loss:  tensor(0.0034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0271, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1846, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0823, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2290, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0073, 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(0.1232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0447, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1077, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4632991318074191, 'recall': 0.5288288288288289, 'f1': 0.4938998737904922, 'number': 1110}, 'P': {'precision': 0.6259157509157509, 'recall': 0.7006663249615582, 'f1': 0.6611850060459493, 'number': 1951}, 'overall_precision': 0.566212691973341, 'overall_recall': 0.6383534792551454, 'overall_f1': 0.6001228501228502, 'overall_accuracy': 0.8281701058483909}
			------------EPOCH 22---------------
Loss:  tensor(0.0030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0240, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1637, 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(0.0026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0710, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2165, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1129, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0310, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0868, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.491928632115548, 'recall': 0.5216216216216216, 'f1': 0.5063401836466986, 'number': 1110}, 'P': {'precision': 0.626912691269127, 'recall': 0.7139928241927217, 'f1': 0.6676252096812845, 'number': 1951}, 'overall_precision': 0.5801706384230656, 'overall_recall': 0.644233910486769, 'overall_f1': 0.6105263157894737, 'overall_accuracy': 0.8303084571795146}
			------------EPOCH 23---------------
Loss:  tensor(0.0028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0240, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1685, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0697, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1153, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0277, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0843, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.46021840873634945, 'recall': 0.5315315315315315, 'f1': 0.49331103678929766, 'number': 1110}, 'P': {'precision': 0.6305762081784386, 'recall': 0.6955407483341876, 'f1': 0.6614672191079697, 'number': 1951}, 'overall_precision': 0.5669772859638905, 'overall_recall': 0.6360666448872917, 'overall_f1': 0.5995381062355658, 'overall_accuracy': 0.8277157061905271}
			------------EPOCH 24---------------
Loss:  tensor(0.0026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0264, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1779, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0779, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2289, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, 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(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1218, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0314, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0793, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4598597038191738, 'recall': 0.5315315315315315, 'f1': 0.4931048892603427, 'number': 1110}, 'P': {'precision': 0.6301433194637078, 'recall': 0.6986160943106099, 'f1': 0.6626154594069033, 'number': 1951}, 'overall_precision': 0.5667440510737086, 'overall_recall': 0.6380267886311662, 'overall_f1': 0.6002766251728907, 'overall_accuracy': 0.8292660109055918}
			------------EPOCH 25---------------
Loss:  tensor(0.0022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0190, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1720, 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(0.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0719, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2144, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0988, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0275, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0794, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.49785038693035255, 'recall': 0.5216216216216216, 'f1': 0.5094588649362077, 'number': 1110}, 'P': {'precision': 0.6245519713261649, 'recall': 0.7145053818554588, 'f1': 0.6665072914176429, 'number': 1951}, 'overall_precision': 0.5811487481590575, 'overall_recall': 0.6445606011107481, 'overall_f1': 0.6112143742255266, 'overall_accuracy': 0.8278092590612638}
			------------EPOCH 26---------------
Loss:  tensor(0.0023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0247, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2187, 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(0.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0900, 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(0.0022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1860, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0027, 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(0.1061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0262, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0849, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.42332613390928725, 'recall': 0.5297297297297298, 'f1': 0.4705882352941177, 'number': 1110}, 'P': {'precision': 0.6082568807339449, 'recall': 0.6796514607893388, 'f1': 0.6419753086419753, 'number': 1951}, 'overall_precision': 0.5362846735780331, 'overall_recall': 0.6252858542959817, 'overall_f1': 0.5773755656108597, 'overall_accuracy': 0.8255372607719448}
			------------EPOCH 27---------------
Loss:  tensor(0.0026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0285, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1968, 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>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0748, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0021, 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(0.1687, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, 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(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0280, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1031, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5012942191544435, 'recall': 0.5234234234234234, 'f1': 0.5121198765976201, 'number': 1110}, 'P': {'precision': 0.6295793758480326, 'recall': 0.7134802665299846, 'f1': 0.6689091782796732, 'number': 1951}, 'overall_precision': 0.585459940652819, 'overall_recall': 0.6445606011107481, 'overall_f1': 0.6135904213963614, 'overall_accuracy': 0.8303886453544317}
			------------EPOCH 28---------------
Loss:  tensor(0.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0212, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1413, 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(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0589, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0001, 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(0.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1517, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0734, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0275, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1120, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.46015625, 'recall': 0.5306306306306307, 'f1': 0.492887029288703, 'number': 1110}, 'P': {'precision': 0.6177145479577788, 'recall': 0.6899026140440799, 'f1': 0.65181598062954, 'number': 1951}, 'overall_precision': 0.559410234171726, 'overall_recall': 0.6321463573995426, 'overall_f1': 0.593558282208589, 'overall_accuracy': 0.8266999893082434}
			------------EPOCH 29---------------
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0157, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1149, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0021, 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(0.0398, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1331, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0020, 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(0.0697, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0265, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0973, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4699278267842823, 'recall': 0.527927927927928, 'f1': 0.4972422571064914, 'number': 1110}, 'P': {'precision': 0.6273179556761647, 'recall': 0.7109174782162994, 'f1': 0.6665064872657376, 'number': 1951}, 'overall_precision': 0.570561017929439, 'overall_recall': 0.6445606011107481, 'overall_f1': 0.6053075625095874, 'overall_accuracy': 0.8297471399550946}
			------------EPOCH 30---------------
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0143, 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(0.1021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0343, 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(0.0016, 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(0.1217, 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(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0684, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0261, 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(0.0843, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4665052461662631, 'recall': 0.5207207207207207, 'f1': 0.49212430821626224, 'number': 1110}, 'P': {'precision': 0.6124551971326165, 'recall': 0.7006663249615582, 'f1': 0.6535978962467129, 'number': 1951}, 'overall_precision': 0.5603572457505042, 'overall_recall': 0.6354132636393336, 'overall_f1': 0.5955296999387629, 'overall_accuracy': 0.8287982465519085}


		-------------RUN 3-----------
			------------EPOCH 1---------------
Loss:  tensor(3.7429, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6454, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3996, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0659, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2504, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9200, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2559, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6557, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8814, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7808, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4747, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6856, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5713, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5996, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1868, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5707, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7892, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7833, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3553, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4290, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4612, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2243060765191298, 'recall': 0.27710843373493976, 'f1': 0.24792703150912107, 'number': 1079}, 'P': {'precision': 0.4268642447418738, 'recall': 0.47805139186295503, 'f1': 0.451010101010101, 'number': 1868}, 'overall_precision': 0.34802919708029195, 'overall_recall': 0.4044791313199864, 'overall_f1': 0.3741368487131199, 'overall_accuracy': 0.7341489478298991}
			------------EPOCH 2---------------
Loss:  tensor(2.2495, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5730, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1566, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1901, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0309, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0631, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1237, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8809, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9325, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4182, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9373, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8966, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0714, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6872, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7470, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4453, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6624, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1391, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7797, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1164, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9332, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0217, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2135102533172497, 'recall': 0.32808155699721964, 'f1': 0.25867738399707707, 'number': 1079}, 'P': {'precision': 0.22551390568319227, 'recall': 0.3993576017130621, 'f1': 0.2882534775888717, 'number': 1868}, 'overall_precision': 0.22150624244865083, 'overall_recall': 0.3732609433322022, 'overall_f1': 0.27802350562365724, 'overall_accuracy': 0.7629192240245506}
			------------EPOCH 3---------------
Loss:  tensor(1.4023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0387, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4998, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4355, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3425, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0799, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2373, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3107, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3275, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4704, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2391, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2807, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9620, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2148, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0748, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3241, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8356, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7884, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6729, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0806, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6862, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.30646022423918845, 'recall': 0.5319740500463392, 'f1': 0.38888888888888884, 'number': 1079}, 'P': {'precision': 0.3612818261633011, 'recall': 0.4405781584582441, 'f1': 0.39700916546068493, 'number': 1868}, 'overall_precision': 0.3365454107443989, 'overall_recall': 0.4740413980318968, 'overall_f1': 0.3936320090166244, 'overall_accuracy': 0.7876068610258659}
			------------EPOCH 4---------------
Loss:  tensor(1.0187, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1827, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0212, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9956, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0606, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0522, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8649, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9915, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3652, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2439, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0698, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9932, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2586, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9077, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7767, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0450, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0787, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6797, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5593, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5547, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7942, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4852, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.37093425605536334, 'recall': 0.49675625579240035, 'f1': 0.4247226624405705, 'number': 1079}, 'P': {'precision': 0.40815533980582525, 'recall': 0.5626338329764454, 'f1': 0.47310375872158456, 'number': 1868}, 'overall_precision': 0.39477611940298507, 'overall_recall': 0.5385137427892772, 'overall_f1': 0.4555762882158748, 'overall_accuracy': 0.8157469311705392}
			------------EPOCH 5---------------
Loss:  tensor(0.7704, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5238, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8879, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6779, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7538, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8434, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0450, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6633, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6832, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2577, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1758, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7193, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7496, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0782, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5910, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8517, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4794, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7622, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0298, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8975, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6686, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4332, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3744, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5983, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2584, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4232209737827715, 'recall': 0.41890639481000924, 'f1': 0.42105263157894735, 'number': 1079}, 'P': {'precision': 0.44617092119866814, 'recall': 0.645610278372591, 'f1': 0.5276744694815139, 'number': 1868}, 'overall_precision': 0.43967117475470696, 'overall_recall': 0.5626060400407193, 'overall_f1': 0.4935992855016374, 'overall_accuracy': 0.8183773564226217}
			------------EPOCH 6---------------
Loss:  tensor(0.5910, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4320, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7575, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6127, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5430, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5617, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0374, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4758, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5639, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2188, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6971, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0244, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3999, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5842, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3681, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6517, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0258, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6629, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4979, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3120, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2697, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3980, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1976, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4115308151093439, 'recall': 0.5755329008341057, 'f1': 0.47990726429675423, 'number': 1079}, 'P': {'precision': 0.5556112224448898, 'recall': 0.5936830835117773, 'f1': 0.5740165631469979, 'number': 1868}, 'overall_precision': 0.49358059914407987, 'overall_recall': 0.5870376654224635, 'overall_f1': 0.536267823930564, 'overall_accuracy': 0.8076090530469092}
			------------EPOCH 7---------------
Loss:  tensor(0.4139, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2921, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5635, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4180, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0197, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3135, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4098, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4799, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3525, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3601, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5141, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5636, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0143, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4708, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3993, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2481, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1463, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5057251908396947, 'recall': 0.4911955514365153, 'f1': 0.4983544898918665, 'number': 1079}, 'P': {'precision': 0.5988960441582337, 'recall': 0.6970021413276232, 'f1': 0.644235526966848, 'number': 1868}, 'overall_precision': 0.5685909373060211, 'overall_recall': 0.6216491347132678, 'overall_f1': 0.5939374290808882, 'overall_accuracy': 0.814692021043402}
			------------EPOCH 8---------------
Loss:  tensor(0.3536, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2386, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2152, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2717, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3412, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0198, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1652, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2384, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1275, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1225, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3412, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2363, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7380, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2648, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3950, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2144, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5139, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4489, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2967, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1212, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1306, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1886, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0847, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.47039740470397406, 'recall': 0.5375347544022243, 'f1': 0.5017301038062284, 'number': 1079}, 'P': {'precision': 0.6001917545541706, 'recall': 0.6702355460385439, 'f1': 0.6332827516439049, 'number': 1868}, 'overall_precision': 0.5519734859897559, 'overall_recall': 0.6216491347132678, 'overall_f1': 0.5847430577721034, 'overall_accuracy': 0.822459995615958}
			------------EPOCH 9---------------
Loss:  tensor(0.1942, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1627, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2825, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1427, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1748, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2190, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1197, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1732, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2807, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2136, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7129, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1787, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1697, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4171, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3804, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0920, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1738, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0675, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.424287856071964, 'recall': 0.5245597775718258, 'f1': 0.469125569830087, 'number': 1079}, 'P': {'precision': 0.42515197568389057, 'recall': 0.5990364025695931, 'f1': 0.4973333333333333, 'number': 1868}, 'overall_precision': 0.4248613212304589, 'overall_recall': 0.5717678995588734, 'overall_f1': 0.4874873426876898, 'overall_accuracy': 0.8162675361683472}
			------------EPOCH 10---------------
Loss:  tensor(0.1186, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1320, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2213, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0913, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1988, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0955, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1445, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0870, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0904, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2545, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1531, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7897, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1384, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2692, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4567, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3557, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0811, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0906, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1316, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0342, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.41385886840432295, 'recall': 0.603336422613531, 'f1': 0.4909502262443439, 'number': 1079}, 'P': {'precision': 0.6276939655172413, 'recall': 0.6236616702355461, 'f1': 0.6256713211600429, 'number': 1868}, 'overall_precision': 0.5296004666083406, 'overall_recall': 0.6162198846284357, 'overall_f1': 0.5696361355081556, 'overall_accuracy': 0.807704953967558}
			------------EPOCH 11---------------
Loss:  tensor(0.0978, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0901, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1593, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0669, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0756, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1511, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0629, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0705, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0580, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0799, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2404, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2154, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6439, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2356, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3313, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0649, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3170, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1990, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2258, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0566, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1483, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1842, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.37721021611001965, 'recall': 0.7117701575532901, 'f1': 0.4930979133226324, 'number': 1079}, 'P': {'precision': 0.6155315085932527, 'recall': 0.5176659528907923, 'f1': 0.5623727827856936, 'number': 1868}, 'overall_precision': 0.48100914887718327, 'overall_recall': 0.5887343060739735, 'overall_f1': 0.5294476655477571, 'overall_accuracy': 0.7561924594476107}
			------------EPOCH 12---------------
Loss:  tensor(0.3218, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1705, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1627, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0845, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2470, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0884, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0793, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2739, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2440, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1850, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3978, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3156, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0087, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2715, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1380, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1890, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1127, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3826227738801943, 'recall': 0.6570898980537535, 'f1': 0.48362892223738063, 'number': 1079}, 'P': {'precision': 0.628385698808234, 'recall': 0.6209850107066381, 'f1': 0.6246634356488961, 'number': 1868}, 'overall_precision': 0.505271695052717, 'overall_recall': 0.6342042755344418, 'overall_f1': 0.5624435750827567, 'overall_accuracy': 0.7680156729504604}
			------------EPOCH 13---------------
Loss:  tensor(0.3212, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2373, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4265, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2339, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2238, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1942, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0984, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0491, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0784, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2338, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1729, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1801, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3563, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0129, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2511, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2454, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1141, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0760, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2222, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1416, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5168107588856868, 'recall': 0.4986098239110287, 'f1': 0.5075471698113208, 'number': 1079}, 'P': {'precision': 0.6175908221797323, 'recall': 0.6916488222698073, 'f1': 0.6525252525252525, 'number': 1868}, 'overall_precision': 0.5841046919885095, 'overall_recall': 0.6209704784526637, 'overall_f1': 0.6019736842105264, 'overall_accuracy': 0.8182677553704516}
			------------EPOCH 14---------------
Loss:  tensor(0.1399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1714, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2439, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2247, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1373, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1393, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0711, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0630, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0948, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2220, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8979, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1356, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2789, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1297, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2828, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0087, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1993, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1971, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0819, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0758, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1322, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0625, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.48795811518324606, 'recall': 0.4318813716404078, 'f1': 0.45821042281219265, 'number': 1079}, 'P': {'precision': 0.5018422991893884, 'recall': 0.7291220556745182, 'f1': 0.5945002182453079, 'number': 1868}, 'overall_precision': 0.49822840010902153, 'overall_recall': 0.6202918221920597, 'overall_f1': 0.5525997581620314, 'overall_accuracy': 0.8124862998684788}
			------------EPOCH 15---------------
Loss:  tensor(0.1293, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1373, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1673, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0735, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1177, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0310, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0419, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0341, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0495, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1940, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1283, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1187, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1897, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2729, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1718, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1745, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0319, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0271, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0598, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0117, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4581589958158996, 'recall': 0.6088971269694161, 'f1': 0.5228810187027457, 'number': 1079}, 'P': {'precision': 0.5893462469733656, 'recall': 0.6514989293361885, 'f1': 0.6188660055936944, 'number': 1868}, 'overall_precision': 0.5355815947413547, 'overall_recall': 0.6359009161859518, 'overall_f1': 0.5814458578963699, 'overall_accuracy': 0.8160620341955283}
			------------EPOCH 16---------------
Loss:  tensor(0.0385, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0631, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0683, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0361, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0687, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0299, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0390, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0417, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0608, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1615, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1237, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7550, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0657, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0208, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1837, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0288, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0303, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0595, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0288, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.38756536897152816, 'recall': 0.6181649675625579, 'f1': 0.4764285714285713, 'number': 1079}, 'P': {'precision': 0.588, 'recall': 0.5508565310492506, 'f1': 0.5688225538971808, 'number': 1868}, 'overall_precision': 0.48861999423797176, 'overall_recall': 0.5755005089921954, 'overall_f1': 0.5285135556248052, 'overall_accuracy': 0.7885110697062692}
			------------EPOCH 17---------------
Loss:  tensor(0.0757, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0979, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1112, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0571, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0694, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0186, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0350, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0714, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1316, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0767, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0724, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1982, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0623, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2570, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1130, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1299, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0238, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0313, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0149, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.527363184079602, 'recall': 0.4911955514365153, 'f1': 0.508637236084453, 'number': 1079}, 'P': {'precision': 0.5368754956383822, 'recall': 0.7248394004282656, 'f1': 0.6168564920273348, 'number': 1868}, 'overall_precision': 0.5341650127587184, 'overall_recall': 0.6392941974889719, 'overall_f1': 0.5820203892493049, 'overall_accuracy': 0.8232409031126698}
			------------EPOCH 18---------------
Loss:  tensor(0.0229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0949, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0603, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0293, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0284, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0230, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0252, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0402, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0493, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1567, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0711, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3987, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0675, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1629, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0268, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2307, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0801, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1121, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0210, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0144, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0044, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4399260628465804, 'recall': 0.6617238183503243, 'f1': 0.5284974093264249, 'number': 1079}, 'P': {'precision': 0.639751552795031, 'recall': 0.6065310492505354, 'f1': 0.6226985435559219, 'number': 1868}, 'overall_precision': 0.5441956393635828, 'overall_recall': 0.6267390566677977, 'overall_f1': 0.5825579561583347, 'overall_accuracy': 0.8050060280578694}
			------------EPOCH 19---------------
Loss:  tensor(0.0172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0392, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0518, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0180, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1111, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, 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(0.0338, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0684, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1794, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0644, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0601, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1823, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0102, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2266, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0633, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0870, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0042, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5103189493433395, 'recall': 0.5041705282669138, 'f1': 0.5072261072261072, 'number': 1079}, 'P': {'precision': 0.575796178343949, 'recall': 0.7259100642398287, 'f1': 0.6421974899360644, 'number': 1868}, 'overall_precision': 0.5553931598947676, 'overall_recall': 0.6447234475738038, 'overall_f1': 0.5967336683417085, 'overall_accuracy': 0.8260494300745287}
			------------EPOCH 20---------------
Loss:  tensor(0.0131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0374, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0370, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0169, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0224, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0144, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0167, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0555, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0634, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1337, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0927, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4252, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0105, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0789, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0908, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0030, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.43133462282398455, 'recall': 0.6200185356811863, 'f1': 0.5087452471482891, 'number': 1079}, 'P': {'precision': 0.6352749047359826, 'recall': 0.6247323340471093, 'f1': 0.6299595141700406, 'number': 1868}, 'overall_precision': 0.5419126328217237, 'overall_recall': 0.6230064472344757, 'overall_f1': 0.5796369376479874, 'overall_accuracy': 0.8072391494958352}
			------------EPOCH 21---------------
Loss:  tensor(0.0097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0387, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0144, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0116, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0893, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0128, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0303, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0684, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1748, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0605, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0836, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2988, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1087, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1085, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0114, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0033, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4727126805778491, 'recall': 0.5458758109360519, 'f1': 0.5066666666666666, 'number': 1079}, 'P': {'precision': 0.605568445475638, 'recall': 0.6986081370449678, 'f1': 0.6487695749440715, 'number': 1868}, 'overall_precision': 0.5568950308732725, 'overall_recall': 0.6426874787919918, 'overall_f1': 0.5967233774417139, 'overall_accuracy': 0.8241725120561157}
			------------EPOCH 22---------------
Loss:  tensor(0.0079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0313, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0139, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0424, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0157, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0331, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3582, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0463, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1236, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1978, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0863, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0030, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4613050075872534, 'recall': 0.5634847080630213, 'f1': 0.5073007926574885, 'number': 1079}, 'P': {'precision': 0.5981264637002341, 'recall': 0.6836188436830836, 'f1': 0.6380214838870846, 'number': 1868}, 'overall_precision': 0.545902114103678, 'overall_recall': 0.6396335256192739, 'overall_f1': 0.5890625000000002, 'overall_accuracy': 0.8253918237615081}
			------------EPOCH 23---------------
Loss:  tensor(0.0066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0284, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0276, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0115, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0320, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0275, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0306, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2883, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0294, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0860, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1436, 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(0.0784, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0076, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0022, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4642324888226528, 'recall': 0.577386468952734, 'f1': 0.5146633622470055, 'number': 1079}, 'P': {'precision': 0.6127618119824647, 'recall': 0.6734475374732334, 'f1': 0.6416730425911757, 'number': 1868}, 'overall_precision': 0.5540500736377025, 'overall_recall': 0.6382762130980658, 'overall_f1': 0.5931882686849574, 'overall_accuracy': 0.8189390618149934}
			------------EPOCH 24---------------
Loss:  tensor(0.0051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0276, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0254, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0237, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0945, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0275, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2480, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0188, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0633, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1152, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0377, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0563, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4461538461538462, 'recall': 0.5644114921223355, 'f1': 0.49836333878887074, 'number': 1079}, 'P': {'precision': 0.5968360498561841, 'recall': 0.6664882226980728, 'f1': 0.6297420333839151, 'number': 1868}, 'overall_precision': 0.5372355838887279, 'overall_recall': 0.6291143535799117, 'overall_f1': 0.5795561112847765, 'overall_accuracy': 0.8203227750986409}
			------------EPOCH 25---------------
Loss:  tensor(0.0045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0267, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0214, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, 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(0.0042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0218, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0956, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2592, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0548, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0308, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0504, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4873675631621842, 'recall': 0.5542168674698795, 'f1': 0.5186470078057243, 'number': 1079}, 'P': {'precision': 0.60398332561371, 'recall': 0.6980728051391863, 'f1': 0.6476285075738764, 'number': 1868}, 'overall_precision': 0.561724748966332, 'overall_recall': 0.6454021038344079, 'overall_f1': 0.600663192799621, 'overall_accuracy': 0.8237067075843928}
			------------EPOCH 26---------------
Loss:  tensor(0.0040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0279, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0090, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0214, 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(0.0037, 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(0.1091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2584, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0540, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0284, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0445, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, 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(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.42827868852459017, 'recall': 0.5810936051899908, 'f1': 0.4931183641368462, 'number': 1079}, 'P': {'precision': 0.6006021073758153, 'recall': 0.6407922912205567, 'f1': 0.62004662004662, 'number': 1868}, 'overall_precision': 0.5276251084755569, 'overall_recall': 0.6189345096708517, 'overall_f1': 0.5696439725171768, 'overall_accuracy': 0.8140618149934239}
			------------EPOCH 27---------------
Loss:  tensor(0.0042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0325, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0212, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2443, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0168, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0571, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1240, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0242, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0392, 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(0.0060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.502116850127011, 'recall': 0.5495829471733086, 'f1': 0.5247787610619468, 'number': 1079}, 'P': {'precision': 0.6088154269972452, 'recall': 0.7098501070663812, 'f1': 0.6554621848739496, 'number': 1868}, 'overall_precision': 0.5713009824352486, 'overall_recall': 0.6511706820495419, 'overall_f1': 0.6086267047256582, 'overall_accuracy': 0.8257754274441035}
			------------EPOCH 28---------------
Loss:  tensor(0.0036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0331, 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(0.0027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0181, 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(0.0039, 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(0.1209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0246, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2275, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0132, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1130, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0240, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0374, 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(0.0059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.44002838892831797, 'recall': 0.5746061167747915, 'f1': 0.4983922829581994, 'number': 1079}, 'P': {'precision': 0.6015586945932782, 'recall': 0.661134903640257, 'f1': 0.6299413414945166, 'number': 1868}, 'overall_precision': 0.5358174465626805, 'overall_recall': 0.6294536817102138, 'overall_f1': 0.5788734591980027, 'overall_accuracy': 0.8191719640508549}
			------------EPOCH 29---------------
Loss:  tensor(0.0035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0310, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0191, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0190, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0155, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0483, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1170, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0248, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0364, 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(0.0057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.47788309636650866, 'recall': 0.5607043558850788, 'f1': 0.5159914712153518, 'number': 1079}, 'P': {'precision': 0.6100331282536677, 'recall': 0.6900428265524625, 'f1': 0.6475759859331827, 'number': 1868}, 'overall_precision': 0.5605208641609943, 'overall_recall': 0.6426874787919918, 'overall_f1': 0.5987986089155864, 'overall_accuracy': 0.8240903112669882}
			------------EPOCH 30---------------
Loss:  tensor(0.0028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0244, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0111, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, 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(0.0022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0123, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0896, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1662, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0213, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1176, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0255, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0503, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4659969088098918, 'recall': 0.5588507877664504, 'f1': 0.5082174462705435, 'number': 1079}, 'P': {'precision': 0.6060606060606061, 'recall': 0.6852248394004282, 'f1': 0.6432160804020101, 'number': 1868}, 'overall_precision': 0.5528479154433353, 'overall_recall': 0.6389548693586699, 'overall_f1': 0.5927908074925232, 'overall_accuracy': 0.8234327049539676}


		-------------RUN 4-----------
			------------EPOCH 1---------------
Loss:  tensor(3.5250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4567, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0601, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0932, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3864, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7733, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7802, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6982, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3941, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6704, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5794, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3379, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5841, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0682, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3522, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1819, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6893, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3225, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2406, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1705, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9295, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9586, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5174, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 1183}, 'P': {'precision': 0.36538461538461536, 'recall': 0.5437365010799136, 'f1': 0.43706597222222215, 'number': 1852}, 'overall_precision': 0.36538461538461536, 'overall_recall': 0.3317957166392092, 'overall_f1': 0.34778103954412015, 'overall_accuracy': 0.6855522232364063}
			------------EPOCH 2---------------
Loss:  tensor(2.0735, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9866, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3545, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9432, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2680, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8495, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8652, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8631, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2862, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2932, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7905, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7941, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6915, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1802, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7396, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5910, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5775, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7222, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4009, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 1183}, 'P': {'precision': 0.3345291479820628, 'recall': 0.6042116630669546, 'f1': 0.43063305753319225, 'number': 1852}, 'overall_precision': 0.33313486156594224, 'overall_recall': 0.3686985172981878, 'overall_f1': 0.35001563966218335, 'overall_accuracy': 0.6994784196114227}
			------------EPOCH 3---------------
Loss:  tensor(1.6257, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1285, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4734, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3960, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4361, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4520, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2336, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3913, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3420, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2750, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3887, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3819, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3904, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4688, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9429, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3137, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2640, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0660, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2926, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5590, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3263, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.1568627450980392, 'recall': 0.027049873203719356, 'f1': 0.046142754145638065, 'number': 1183}, 'P': {'precision': 0.3797909407665505, 'recall': 0.6474082073434125, 'f1': 0.47873827111199846, 'number': 1852}, 'overall_precision': 0.36626004165426956, 'overall_recall': 0.4056013179571664, 'overall_f1': 0.38492808005003126, 'overall_accuracy': 0.7199243708436562}
			------------EPOCH 4---------------
Loss:  tensor(1.2673, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2536, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1475, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1532, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1378, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1198, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1641, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1422, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1760, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0698, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0301, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2509, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0483, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0970, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0629, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1380, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2121, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7296, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0306, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0498, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0930, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3623, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2715, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4045346062052506, 'recall': 0.2865595942519019, 'f1': 0.3354774863928748, 'number': 1183}, 'P': {'precision': 0.4390763765541741, 'recall': 0.6673866090712743, 'f1': 0.5296764516820228, 'number': 1852}, 'overall_precision': 0.43115247741582263, 'overall_recall': 0.5189456342668863, 'overall_f1': 0.4709928229665072, 'overall_accuracy': 0.7820706741426522}
			------------EPOCH 5---------------
Loss:  tensor(1.0151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9872, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6538, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0423, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1147, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9164, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8329, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8295, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8459, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1379, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8454, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6970, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2397, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8305, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7728, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7521, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9484, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0483, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5251, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8107, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6946, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0334, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8865, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2761, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2135, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.43050039714058774, 'recall': 0.45815722738799664, 'f1': 0.4438984438984439, 'number': 1183}, 'P': {'precision': 0.5324173465006441, 'recall': 0.6695464362850972, 'f1': 0.5931595312126287, 'number': 1852}, 'overall_precision': 0.49665551839464883, 'overall_recall': 0.5871499176276771, 'overall_f1': 0.5381247168956665, 'overall_accuracy': 0.8051114878080584}
			------------EPOCH 6---------------
Loss:  tensor(0.8649, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8149, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4295, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7407, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0950, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6720, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6400, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6331, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5353, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6290, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5498, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7386, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8220, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3331, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6242, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5674, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0316, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7483, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2332, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1720, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.478515625, 'recall': 0.20710059171597633, 'f1': 0.2890855457227139, 'number': 1183}, 'P': {'precision': 0.5073260073260073, 'recall': 0.7478401727861771, 'f1': 0.6045395024006984, 'number': 1852}, 'overall_precision': 0.5027760641579272, 'overall_recall': 0.5370675453047776, 'overall_f1': 0.5193563804365143, 'overall_accuracy': 0.7694093102099361}
			------------EPOCH 7---------------
Loss:  tensor(0.8547, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7626, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2828, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6115, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0576, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5155, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4696, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4473, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4321, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0789, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4960, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3755, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1714, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4213, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3223, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5797, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2363, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4532, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0164, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6839, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1515, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1790, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5085227272727273, 'recall': 0.4539306846999155, 'f1': 0.47967842786958464, 'number': 1183}, 'P': {'precision': 0.5764172335600907, 'recall': 0.6862850971922246, 'f1': 0.6265713581464136, 'number': 1852}, 'overall_precision': 0.5544311560870898, 'overall_recall': 0.5957166392092257, 'overall_f1': 0.5743329097839899, 'overall_accuracy': 0.8147346459773113}
			------------EPOCH 8---------------
Loss:  tensor(0.6170, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6265, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1821, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5187, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0705, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3925, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3441, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3978, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0631, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3413, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3130, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1481, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3553, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2461, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4786, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7343, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2308, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4464, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2734, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0144, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5904, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1636, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5187878787878788, 'recall': 0.3617920540997464, 'f1': 0.42629482071713154, 'number': 1183}, 'P': {'precision': 0.5967169476486247, 'recall': 0.7262419006479481, 'f1': 0.6551388212372138, 'number': 1852}, 'overall_precision': 0.575836310490419, 'overall_recall': 0.5841845140032949, 'overall_f1': 0.5799803729146221, 'overall_accuracy': 0.801590820185161}
			------------EPOCH 9---------------
Loss:  tensor(0.4960, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5130, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1445, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5325, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0387, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2763, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1747, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2187, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1825, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0497, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1373, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2951, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5133, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6804, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1686, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3486, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1829, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5925, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0956, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1982, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.41098677517802645, 'recall': 0.3415046491969569, 'f1': 0.3730378578024007, 'number': 1183}, 'P': {'precision': 0.5032751091703057, 'recall': 0.7467602591792657, 'f1': 0.601304347826087, 'number': 1852}, 'overall_precision': 0.478960064325918, 'overall_recall': 0.5887973640856672, 'overall_f1': 0.5282293822051434, 'overall_accuracy': 0.7731646890076933}
			------------EPOCH 10---------------
Loss:  tensor(0.6047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4755, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1234, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4893, 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(0.1400, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2190, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2076, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0340, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3340, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1107, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1667, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4592, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7443, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1371, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2171, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0148, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0978, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1465, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.49951314508276534, 'recall': 0.43364327979712597, 'f1': 0.46425339366515833, 'number': 1183}, 'P': {'precision': 0.5492473118279569, 'recall': 0.6895248380129589, 'f1': 0.6114436198228393, 'number': 1852}, 'overall_precision': 0.5340095465393795, 'overall_recall': 0.5897858319604613, 'overall_f1': 0.560513543134492, 'overall_accuracy': 0.7932455339679229}
			------------EPOCH 11---------------
Loss:  tensor(0.5299, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1352, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0269, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2774, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1527, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1885, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1521, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0243, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1976, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1203, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1941, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1711, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1710, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3897, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5803, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1891, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2923, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2467, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4534, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1162, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5154061624649859, 'recall': 0.46661031276415893, 'f1': 0.48979591836734687, 'number': 1183}, 'P': {'precision': 0.6218407248450167, 'recall': 0.7041036717062635, 'f1': 0.660420359584705, 'number': 1852}, 'overall_precision': 0.5858585858585859, 'overall_recall': 0.6115321252059308, 'overall_f1': 0.5984201192971144, 'overall_accuracy': 0.8151258312687443}
			------------EPOCH 12---------------
Loss:  tensor(0.2947, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2823, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0422, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3520, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1835, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1341, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0207, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2560, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1242, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0761, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1114, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0936, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0750, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2553, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3978, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0734, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1594, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1204, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0105, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4584, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1800, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3671625929861849, 'recall': 0.5841081994928149, 'f1': 0.4508972267536705, 'number': 1183}, 'P': {'precision': 0.4811858608893957, 'recall': 0.4557235421166307, 'f1': 0.46810870770937335, 'number': 1852}, 'overall_precision': 0.4221672167216722, 'overall_recall': 0.5057660626029654, 'overall_f1': 0.4602008694348674, 'overall_accuracy': 0.7702699178510888}
			------------EPOCH 13---------------
Loss:  tensor(0.5550, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0606, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4186, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0084, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0624, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1418, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0942, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0334, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5423, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2163, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1705, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1418, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1277, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0722, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3286, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5325, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0683, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1875, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0225, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5824, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0966, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1664, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4124263524370648, 'recall': 0.650887573964497, 'f1': 0.5049180327868852, 'number': 1183}, 'P': {'precision': 0.667332002661344, 'recall': 0.5415766738660908, 'f1': 0.5979135618479882, 'number': 1852}, 'overall_precision': 0.5261127596439169, 'overall_recall': 0.5841845140032949, 'overall_f1': 0.5536299765807963, 'overall_accuracy': 0.7909766592776112}
			------------EPOCH 14---------------
Loss:  tensor(0.2604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0993, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4105, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0114, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1698, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0567, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1318, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0608, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2208, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1140, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1170, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2165, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4506, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2860, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4364, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6418, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0415, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1230, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0446, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3553, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0537, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2223, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3278183916891112, 'recall': 0.7202028740490279, 'f1': 0.45055526176626126, 'number': 1183}, 'P': {'precision': 0.6421800947867299, 'recall': 0.29265658747300216, 'f1': 0.4020771513353116, 'number': 1852}, 'overall_precision': 0.404879465582341, 'overall_recall': 0.45930807248764416, 'overall_f1': 0.430379746835443, 'overall_accuracy': 0.669044203937932}
			------------EPOCH 15---------------
Loss:  tensor(1.0336, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1833, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6647, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0239, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2695, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1324, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1488, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0686, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0763, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0800, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2437, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4697, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0510, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0286, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6567, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6234, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3323, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0171, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0811, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1117, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.39517241379310347, 'recall': 0.48436179205409974, 'f1': 0.4352449677174326, 'number': 1183}, 'P': {'precision': 0.5672310756972112, 'recall': 0.6150107991360692, 'f1': 0.5901554404145078, 'number': 1852}, 'overall_precision': 0.4950838635049161, 'overall_recall': 0.5640856672158154, 'overall_f1': 0.5273371322963191, 'overall_accuracy': 0.8066762289737905}
			------------EPOCH 16---------------
Loss:  tensor(0.2486, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3289, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0415, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4393, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2439, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2573, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2880, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2403, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1890, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0610, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1823, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1197, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0740, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2819, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4827, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0561, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1535, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3839, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0518, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1028, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5092592592592593, 'recall': 0.4184277261200338, 'f1': 0.4593967517401392, 'number': 1183}, 'P': {'precision': 0.5782145782145782, 'recall': 0.7624190064794817, 'f1': 0.6576618537494179, 'number': 1852}, 'overall_precision': 0.5585823081429409, 'overall_recall': 0.6283360790774299, 'overall_f1': 0.5914095208559467, 'overall_accuracy': 0.8138088407875864}
			------------EPOCH 17---------------
Loss:  tensor(0.2099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2480, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3185, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0090, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0715, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0437, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0546, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0408, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0340, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0578, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0459, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3295, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0542, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0858, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0328, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3105, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0297, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0830, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4959677419354839, 'recall': 0.5198647506339814, 'f1': 0.5076351630210483, 'number': 1183}, 'P': {'precision': 0.5910143584993053, 'recall': 0.6889848812095032, 'f1': 0.6362503116429818, 'number': 1852}, 'overall_precision': 0.5563401000294205, 'overall_recall': 0.6230642504118616, 'overall_f1': 0.5878147342244328, 'overall_accuracy': 0.8256356760985787}
			------------EPOCH 18---------------
Loss:  tensor(0.1112, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1789, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0137, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2358, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0560, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0312, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0638, 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(0.0940, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0333, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0445, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0455, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1819, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3149, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0289, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0623, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0168, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2654, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0166, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0709, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4689221085759245, 'recall': 0.503803888419273, 'f1': 0.4857375713121434, 'number': 1183}, 'P': {'precision': 0.5959221501390176, 'recall': 0.6943844492440605, 'f1': 0.6413965087281797, 'number': 1852}, 'overall_precision': 0.5488480606590843, 'overall_recall': 0.6200988467874794, 'overall_f1': 0.5823019801980198, 'overall_accuracy': 0.8252575303168601}
			------------EPOCH 19---------------
Loss:  tensor(0.0829, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1510, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0085, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0400, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0107, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0670, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0170, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0338, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0126, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1254, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2598, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0195, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0518, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2148, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0666, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5025553662691652, 'recall': 0.4987320371935757, 'f1': 0.5006364022061943, 'number': 1183}, 'P': {'precision': 0.6197449220595181, 'recall': 0.7084233261339092, 'f1': 0.6611237087427564, 'number': 1852}, 'overall_precision': 0.577939835916135, 'overall_recall': 0.6266886326194399, 'overall_f1': 0.6013278533038254, 'overall_accuracy': 0.826183335506585}
			------------EPOCH 20---------------
Loss:  tensor(0.0638, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1386, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1850, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0299, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0382, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0120, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0308, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0126, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2404, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0169, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0434, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1963, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0077, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0603, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5008319467554077, 'recall': 0.5088757396449705, 'f1': 0.5048218029350106, 'number': 1183}, 'P': {'precision': 0.6113941639647985, 'recall': 0.712742980561555, 'f1': 0.6581899775617054, 'number': 1852}, 'overall_precision': 0.571853614995537, 'overall_recall': 0.6332784184514003, 'overall_f1': 0.6010006253908693, 'overall_accuracy': 0.8271221802060242}
			------------EPOCH 21---------------
Loss:  tensor(0.0488, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1177, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1571, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0506, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0283, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0965, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2307, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0421, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1843, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0545, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5004101722723544, 'recall': 0.5156382079459002, 'f1': 0.5079100749375519, 'number': 1183}, 'P': {'precision': 0.6238140417457305, 'recall': 0.7100431965442765, 'f1': 0.6641414141414143, 'number': 1852}, 'overall_precision': 0.5785993387436129, 'overall_recall': 0.6342668863261944, 'overall_f1': 0.6051556114429425, 'overall_accuracy': 0.8269917851088799}
			------------EPOCH 22---------------
Loss:  tensor(0.0420, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1470, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0203, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0297, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0449, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0087, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0259, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0886, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0132, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0367, 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(0.1687, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0535, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4987593052109181, 'recall': 0.5097210481825867, 'f1': 0.5041806020066889, 'number': 1183}, 'P': {'precision': 0.6148459383753502, 'recall': 0.7111231101511879, 'f1': 0.6594892338507762, 'number': 1852}, 'overall_precision': 0.5729632945389436, 'overall_recall': 0.6326194398682042, 'overall_f1': 0.6013153773880363, 'overall_accuracy': 0.827382970400313}
			------------EPOCH 23---------------
Loss:  tensor(0.0354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0960, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0181, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0237, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0903, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2381, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0397, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1515, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0463, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5071010860484545, 'recall': 0.5131022823330516, 'f1': 0.5100840336134453, 'number': 1183}, 'P': {'precision': 0.6143122676579925, 'recall': 0.7138228941684666, 'f1': 0.6603396603396604, 'number': 1852}, 'overall_precision': 0.5759928336816961, 'overall_recall': 0.6355848434925865, 'overall_f1': 0.6043233082706767, 'overall_accuracy': 0.8276959186334594}
			------------EPOCH 24---------------
Loss:  tensor(0.0289, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0847, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0019, 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(0.0171, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0278, 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(0.0406, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0357, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0771, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1665, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0699, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.47557251908396947, 'recall': 0.5266272189349113, 'f1': 0.4997994384275973, 'number': 1183}, 'P': {'precision': 0.6070921985815603, 'recall': 0.693304535637149, 'f1': 0.6473405596168389, 'number': 1852}, 'overall_precision': 0.5567883211678832, 'overall_recall': 0.6283360790774299, 'overall_f1': 0.5904024767801858, 'overall_accuracy': 0.8268092319728778}
			------------EPOCH 25---------------
Loss:  tensor(0.0569, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1959, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0185, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0284, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0246, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1265, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2654, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0102, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0683, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1700, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0503, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4987971130713713, 'recall': 0.525781910397295, 'f1': 0.511934156378601, 'number': 1183}, 'P': {'precision': 0.6116092496460594, 'recall': 0.6997840172786177, 'f1': 0.6527323092420044, 'number': 1852}, 'overall_precision': 0.5698158051099228, 'overall_recall': 0.6319604612850083, 'overall_f1': 0.5992813622871427, 'overall_accuracy': 0.8272525753031686}
			------------EPOCH 26---------------
Loss:  tensor(0.0329, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0885, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1391, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0196, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0329, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0390, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0436, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1390, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3223, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0415, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1644, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0492, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5092281879194631, 'recall': 0.5131022823330516, 'f1': 0.5111578947368421, 'number': 1183}, 'P': {'precision': 0.607852193995381, 'recall': 0.7105831533477321, 'f1': 0.6552153348269852, 'number': 1852}, 'overall_precision': 0.5728328865058088, 'overall_recall': 0.6336079077429984, 'overall_f1': 0.6016896120150188, 'overall_accuracy': 0.8288564349980441}
			------------EPOCH 27---------------
Loss:  tensor(0.0412, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0928, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1367, 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(0.0028, 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(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0321, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0777, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1819, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0303, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1413, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0427, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5122777307366638, 'recall': 0.5114116652578191, 'f1': 0.511844331641286, 'number': 1183}, 'P': {'precision': 0.6170411985018727, 'recall': 0.7116630669546437, 'f1': 0.6609829488465396, 'number': 1852}, 'overall_precision': 0.5797407295749171, 'overall_recall': 0.6336079077429984, 'overall_f1': 0.6054785894206549, 'overall_accuracy': 0.8288433954883296}
			------------EPOCH 28---------------
Loss:  tensor(0.0367, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0825, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1150, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0177, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0199, 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(0.0274, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0244, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0625, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1677, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0241, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1286, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0373, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.49551752241238795, 'recall': 0.5139475908706678, 'f1': 0.5045643153526971, 'number': 1183}, 'P': {'precision': 0.6022988505747127, 'recall': 0.7073434125269978, 'f1': 0.6506083933449218, 'number': 1852}, 'overall_precision': 0.5637860082304527, 'overall_recall': 0.6319604612850083, 'overall_f1': 0.5959297809538605, 'overall_accuracy': 0.8291172251923328}
			------------EPOCH 29---------------
Loss:  tensor(0.0223, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0693, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0972, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0116, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0148, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0249, 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(0.0025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0491, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1615, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0063, 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(0.0001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0301, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.49594155844155846, 'recall': 0.5164835164835165, 'f1': 0.5060041407867495, 'number': 1183}, 'P': {'precision': 0.6120527306967984, 'recall': 0.7019438444924406, 'f1': 0.6539235412474849, 'number': 1852}, 'overall_precision': 0.5694278903456496, 'overall_recall': 0.6296540362438221, 'overall_f1': 0.5980284775465498, 'overall_accuracy': 0.8284652497066111}
			------------EPOCH 30---------------
Loss:  tensor(0.0147, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0536, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0795, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0130, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0207, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0200, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0450, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1532, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0209, 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(0.0967, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0266, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5062344139650873, 'recall': 0.514792899408284, 'f1': 0.5104777870913663, 'number': 1183}, 'P': {'precision': 0.6042632066728453, 'recall': 0.7041036717062635, 'f1': 0.6503740648379053, 'number': 1852}, 'overall_precision': 0.5691758405236537, 'overall_recall': 0.6303130148270181, 'overall_f1': 0.5981863664790493, 'overall_accuracy': 0.8279827878471769}


		-------------RUN 5-----------
			------------EPOCH 1---------------
Loss:  tensor(1.4514, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5831, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3958, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3511, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0769, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0507, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5939, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9293, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7501, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6484, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3407, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5732, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5361, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7830, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2758, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3976, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3341, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2664, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2434, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1952, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2273, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9560, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2839, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.07547169811320754, 'recall': 0.011560693641618497, 'f1': 0.020050125313283207, 'number': 1038}, 'P': {'precision': 0.3873486919172198, 'recall': 0.5447556287753982, 'f1': 0.4527612962117754, 'number': 1821}, 'overall_precision': 0.36911764705882355, 'overall_recall': 0.35117173837005944, 'overall_f1': 0.3599211328195017, 'overall_accuracy': 0.6814441956965022}
			------------EPOCH 2---------------
Loss:  tensor(0.8573, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9938, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3320, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9370, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9284, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6936, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8860, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7340, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3955, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6294, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2273, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5658, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5144, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1427, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5305, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5752, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5695, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5534, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6204, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4496, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6602, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6459, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3063, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.18887908208296558, 'recall': 0.20616570327552985, 'f1': 0.1971441731920774, 'number': 1038}, 'P': {'precision': 0.34469551895825357, 'recall': 0.4942339373970346, 'f1': 0.4061371841155235, 'number': 1821}, 'overall_precision': 0.29754273504273504, 'overall_recall': 0.38964672962574326, 'overall_f1': 0.3374223837649553, 'overall_accuracy': 0.741918464543912}
			------------EPOCH 3---------------
Loss:  tensor(0.5861, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3382, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8748, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2979, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3329, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1957, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5155, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2806, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2242, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3112, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2243, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1531, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0986, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8176, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1809, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2503, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0880, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1771, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2673, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2317, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0890, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2895, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4522, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2593, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.46026986506746626, 'recall': 0.2957610789980732, 'f1': 0.36011730205278586, 'number': 1038}, 'P': {'precision': 0.39840499306518723, 'recall': 0.6309719934102141, 'f1': 0.4884165781083952, 'number': 1821}, 'overall_precision': 0.410025344973247, 'overall_recall': 0.509268975166142, 'overall_f1': 0.45429017160686425, 'overall_accuracy': 0.7833278672161302}
			------------EPOCH 4---------------
Loss:  tensor(0.4440, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9963, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6843, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0279, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9213, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9587, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9760, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2616, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8703, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0994, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8456, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7430, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6615, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9649, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9410, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8465, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8699, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9668, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9885, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1154, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2402, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.40634920634920635, 'recall': 0.3699421965317919, 'f1': 0.3872919818456884, 'number': 1038}, 'P': {'precision': 0.4344109996283909, 'recall': 0.641954969796815, 'f1': 0.5181737588652483, 'number': 1821}, 'overall_precision': 0.42711771177117713, 'overall_recall': 0.5431969220006996, 'overall_f1': 0.47821401077752124, 'overall_accuracy': 0.8080252106831696}
			------------EPOCH 5---------------
Loss:  tensor(0.3514, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7319, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4812, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7797, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8240, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6966, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0527, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7295, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2115, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7662, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0824, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8274, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7555, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4629, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6876, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7942, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6889, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9000, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9908, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7199, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8752, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2494, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1882, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.32466124661246615, 'recall': 0.5770712909441233, 'f1': 0.4155393687131461, 'number': 1038}, 'P': {'precision': 0.5078625509609784, 'recall': 0.47885777045579353, 'f1': 0.49293386093838326, 'number': 1821}, 'overall_precision': 0.41297024143739475, 'overall_recall': 0.5145155648828261, 'overall_f1': 0.4581840834760941, 'overall_accuracy': 0.7804474611073878}
			------------EPOCH 6---------------
Loss:  tensor(0.2793, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6609, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4150, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7378, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5798, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5655, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9377, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5908, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5644, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1613, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5939, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0539, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5224, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4520, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3947, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5532, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4608, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5370, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4236, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7860, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1828, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1479, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.48523622047244097, 'recall': 0.4749518304431599, 'f1': 0.48003894839337874, 'number': 1038}, 'P': {'precision': 0.6381260096930533, 'recall': 0.6507413509060955, 'f1': 0.6443719412724306, 'number': 1821}, 'overall_precision': 0.5840584754611904, 'overall_recall': 0.5869185029730675, 'overall_f1': 0.5854849965108164, 'overall_accuracy': 0.8094939326099047}
			------------EPOCH 7---------------
Loss:  tensor(0.1726, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4498, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3204, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4786, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3882, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4000, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8107, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3808, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1085, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3324, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0270, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3434, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3345, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2903, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3956, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3597, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3747, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6404, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5474, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3255, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5663, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1233, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1028, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.45507246376811594, 'recall': 0.45375722543352603, 'f1': 0.4544138929088278, 'number': 1038}, 'P': {'precision': 0.5967402733964248, 'recall': 0.6232839099395936, 'f1': 0.6097233413913511, 'number': 1821}, 'overall_precision': 0.5468164794007491, 'overall_recall': 0.5617348723329836, 'overall_f1': 0.5541752933057281, 'overall_accuracy': 0.7926392790429067}
			------------EPOCH 8---------------
Loss:  tensor(0.1280, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2718, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1948, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3569, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2797, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2660, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2789, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2563, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0883, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2535, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0198, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2693, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2276, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1896, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2800, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3298, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1819, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2314, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5548, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3335, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1857, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4942, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0950, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0940, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3631578947368421, 'recall': 0.4653179190751445, 'f1': 0.40793918918918926, 'number': 1038}, 'P': {'precision': 0.45993802567507747, 'recall': 0.5705656232839099, 'f1': 0.5093137254901962, 'number': 1821}, 'overall_precision': 0.42407355809417663, 'overall_recall': 0.5323539699195523, 'overall_f1': 0.47208436724565755, 'overall_accuracy': 0.7974447090362047}
			------------EPOCH 9---------------
Loss:  tensor(0.0988, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1949, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1640, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2164, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1651, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2224, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0727, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1611, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0129, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1922, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1788, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1610, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3513, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2193, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1827, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3948, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3320, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1214, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3631, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0809, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0581, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.45193929173693087, 'recall': 0.5163776493256262, 'f1': 0.48201438848920863, 'number': 1038}, 'P': {'precision': 0.5175879396984925, 'recall': 0.6221856123009335, 'f1': 0.5650872817955112, 'number': 1821}, 'overall_precision': 0.4945185185185185, 'overall_recall': 0.583770549143057, 'overall_f1': 0.535450753930061, 'overall_accuracy': 0.8117611829628256}
			------------EPOCH 10---------------
Loss:  tensor(0.0512, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1144, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1504, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1787, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2721, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2318, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7713, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3623, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2130, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1446, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1541, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1318, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2794, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1180, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1585, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4344, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1901, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1289, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1507, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.48947849954254347, 'recall': 0.5154142581888247, 'f1': 0.5021116846550915, 'number': 1038}, 'P': {'precision': 0.6244520214320507, 'recall': 0.7040087863811093, 'f1': 0.6618482188951987, 'number': 1821}, 'overall_precision': 0.5775588048315321, 'overall_recall': 0.6355369010143407, 'overall_f1': 0.6051623646960865, 'overall_accuracy': 0.8159819760726661}
			------------EPOCH 11---------------
Loss:  tensor(0.0810, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2391, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1560, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2777, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1187, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1493, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1910, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0642, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2615, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0236, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1767, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3190, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1401, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1510, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2169, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1076, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2965, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2588, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0797, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0574, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0619, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4693140794223827, 'recall': 0.5009633911368016, 'f1': 0.4846225535880708, 'number': 1038}, 'P': {'precision': 0.5787852112676056, 'recall': 0.7221306974190006, 'f1': 0.6425604690935744, 'number': 1821}, 'overall_precision': 0.5428994082840237, 'overall_recall': 0.6418328086743617, 'overall_f1': 0.5882352941176471, 'overall_accuracy': 0.7979723081749348}
			------------EPOCH 12---------------
Loss:  tensor(0.0444, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3766, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2753, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1758, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1734, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0527, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0077, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1460, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1463, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1478, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2883, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1255, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2571, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1467, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2157, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0486, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0600, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4412081984897519, 'recall': 0.3940269749518304, 'f1': 0.41628498727735364, 'number': 1038}, 'P': {'precision': 0.5771355716107097, 'recall': 0.7457440966501923, 'f1': 0.6506947771921417, 'number': 1821}, 'overall_precision': 0.5387195121951219, 'overall_recall': 0.6180482686253935, 'overall_f1': 0.5756637888906988, 'overall_accuracy': 0.8078683568851688}
			------------EPOCH 13---------------
Loss:  tensor(0.0420, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1272, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0905, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2318, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1571, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1711, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7885, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1562, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1742, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1390, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1249, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0791, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0975, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1356, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0588, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0895, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2823, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2153, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2204, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0616, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0854, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5538641686182669, 'recall': 0.4556840077071291, 'f1': 0.5, 'number': 1038}, 'P': {'precision': 0.5532271584241408, 'recall': 0.7248764415156508, 'f1': 0.6275255526503447, 'number': 1821}, 'overall_precision': 0.553395061728395, 'overall_recall': 0.627142357467646, 'overall_f1': 0.5879652402033119, 'overall_accuracy': 0.8209157409916011}
			------------EPOCH 14---------------
Loss:  tensor(0.0215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0864, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0660, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0874, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0828, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1429, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0433, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0702, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0115, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0430, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0692, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0548, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0623, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0539, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0488, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2787, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1531, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0379, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1705, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0419, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0530, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.47899910634495085, 'recall': 0.5163776493256262, 'f1': 0.49698655540102, 'number': 1038}, 'P': {'precision': 0.6129032258064516, 'recall': 0.6781987918725975, 'f1': 0.6438998957247133, 'number': 1821}, 'overall_precision': 0.5650925335035099, 'overall_recall': 0.6194473592165093, 'overall_f1': 0.5910228600033371, 'overall_accuracy': 0.8207446277574185}
			------------EPOCH 15---------------
Loss:  tensor(0.0147, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0326, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0414, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0279, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3944, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0799, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1397, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0391, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1241, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0528, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0336, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0374, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0716, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0140, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1703, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1878, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0168, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1329, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0163, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0671, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4332688588007737, 'recall': 0.6473988439306358, 'f1': 0.5191193511008111, 'number': 1038}, 'P': {'precision': 0.6664619164619164, 'recall': 0.5958264689730917, 'f1': 0.6291678747463032, 'number': 1821}, 'overall_precision': 0.5526895250078641, 'overall_recall': 0.6145505421476041, 'overall_f1': 0.5819807883405101, 'overall_accuracy': 0.8046457243080608}
			------------EPOCH 16---------------
Loss:  tensor(0.0119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0621, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1180, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0614, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0348, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0482, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5127, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0804, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0756, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0424, 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(0.0231, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0373, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0370, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0507, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0759, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0470, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0530, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2943, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1328, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0230, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0399, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4621710526315789, 'recall': 0.5414258188824663, 'f1': 0.4986690328305235, 'number': 1038}, 'P': {'precision': 0.6160670280926565, 'recall': 0.686436024162548, 'f1': 0.6493506493506493, 'number': 1821}, 'overall_precision': 0.5583975346687211, 'overall_recall': 0.633788037775446, 'overall_f1': 0.5937090432503278, 'overall_accuracy': 0.8222133496841535}
			------------EPOCH 17---------------
Loss:  tensor(0.0051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0220, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0672, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0279, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3367, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0495, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0707, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0333, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0188, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0387, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0254, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0187, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0865, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0168, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0157, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1639, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1240, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0932, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0362, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.48142031379025596, 'recall': 0.5616570327552987, 'f1': 0.5184526456202756, 'number': 1038}, 'P': {'precision': 0.6272219400711021, 'recall': 0.6781987918725975, 'f1': 0.6517150395778364, 'number': 1821}, 'overall_precision': 0.5716981132075472, 'overall_recall': 0.6358866736621196, 'overall_f1': 0.6020864381520119, 'overall_accuracy': 0.8182920047341328}
			------------EPOCH 18---------------
Loss:  tensor(0.0033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0199, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0406, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0174, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0188, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0167, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2902, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0484, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0422, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0262, 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>)
Loss:  tensor(0.0089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0223, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0345, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1266, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1479, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0715, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0324, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.496551724137931, 'recall': 0.5549132947976878, 'f1': 0.5241128298453139, 'number': 1038}, 'P': {'precision': 0.6425641025641026, 'recall': 0.6880834706205382, 'f1': 0.6645452134712277, 'number': 1821}, 'overall_precision': 0.5881028938906753, 'overall_recall': 0.639734172787688, 'overall_f1': 0.6128329703467917, 'overall_accuracy': 0.8223416846097905}
			------------EPOCH 19---------------
Loss:  tensor(0.0024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0364, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0114, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0132, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0123, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2610, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0339, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0328, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0225, 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(0.0055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0077, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0241, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0971, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0625, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0290, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.49699054170249357, 'recall': 0.5568400770712909, 'f1': 0.5252158109950023, 'number': 1038}, 'P': {'precision': 0.6266397578203835, 'recall': 0.6820428336079077, 'f1': 0.6531685511438338, 'number': 1821}, 'overall_precision': 0.5786963434022258, 'overall_recall': 0.6365862189576775, 'overall_f1': 0.6062624916722186, 'overall_accuracy': 0.8211724108428753}
			------------EPOCH 20---------------
Loss:  tensor(0.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0474, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0090, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0114, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2374, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0299, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0266, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0202, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0111, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0222, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0915, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0529, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0272, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.48102310231023104, 'recall': 0.5616570327552987, 'f1': 0.5182222222222223, 'number': 1038}, 'P': {'precision': 0.6371134020618556, 'recall': 0.6787479406919276, 'f1': 0.6572720021270939, 'number': 1821}, 'overall_precision': 0.5770939086294417, 'overall_recall': 0.6362364463098986, 'overall_f1': 0.6052237564465148, 'overall_accuracy': 0.8210583353534201}
			------------EPOCH 21---------------
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0105, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0406, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0220, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0190, 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(0.0032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0190, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0836, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0489, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0256, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4978241949521323, 'recall': 0.5510597302504817, 'f1': 0.5230909922267947, 'number': 1038}, 'P': {'precision': 0.6300751879699248, 'recall': 0.6902800658978583, 'f1': 0.6588050314465409, 'number': 1821}, 'overall_precision': 0.5817430025445293, 'overall_recall': 0.639734172787688, 'overall_f1': 0.6093619856738297, 'overall_accuracy': 0.8233683640148869}
			------------EPOCH 22---------------
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0434, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0085, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2000, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0186, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0166, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0922, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0251, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4745484400656814, 'recall': 0.5568400770712909, 'f1': 0.5124113475177304, 'number': 1038}, 'P': {'precision': 0.6424870466321243, 'recall': 0.6809445359692476, 'f1': 0.6611570247933884, 'number': 1821}, 'overall_precision': 0.5775095298602287, 'overall_recall': 0.6358866736621196, 'overall_f1': 0.6052938238721493, 'overall_accuracy': 0.8217142694177872}
			------------EPOCH 23---------------
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0421, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1837, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0211, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0165, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0176, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0154, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0644, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1112, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0235, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5, 'recall': 0.5481695568400771, 'f1': 0.5229779411764706, 'number': 1038}, 'P': {'precision': 0.6242544731610338, 'recall': 0.6897309170785283, 'f1': 0.6553613357683277, 'number': 1821}, 'overall_precision': 0.5793650793650794, 'overall_recall': 0.6383350821965722, 'overall_f1': 0.6074222000332835, 'overall_accuracy': 0.824951161431077}
			------------EPOCH 24---------------
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0383, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0077, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1783, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0140, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0168, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0590, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0897, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0346, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0220, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.47611064543168485, 'recall': 0.5472061657032755, 'f1': 0.5091887046167638, 'number': 1038}, 'P': {'precision': 0.632808532249873, 'recall': 0.6842394288852279, 'f1': 0.6575197889182057, 'number': 1821}, 'overall_precision': 0.5736875395319418, 'overall_recall': 0.6344875830710038, 'overall_f1': 0.6025577146653379, 'overall_accuracy': 0.8224415006630638}
			------------EPOCH 25---------------
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0076, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0403, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1660, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0135, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0128, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0031, 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(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0497, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0286, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0212, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4740313272877164, 'recall': 0.5539499036608864, 'f1': 0.5108840515326522, 'number': 1038}, 'P': {'precision': 0.6424870466321243, 'recall': 0.6809445359692476, 'f1': 0.6611570247933884, 'number': 1821}, 'overall_precision': 0.5774737511931276, 'overall_recall': 0.6348373557187827, 'overall_f1': 0.6047984005331556, 'overall_accuracy': 0.8215146373112407}
			------------EPOCH 26---------------
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0395, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1701, 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(0.0158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0552, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0315, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0297, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5054844606946983, 'recall': 0.5327552986512524, 'f1': 0.5187617260787992, 'number': 1038}, 'P': {'precision': 0.613384689455946, 'recall': 0.6996155958264689, 'f1': 0.6536685479733196, 'number': 1821}, 'overall_precision': 0.5761589403973509, 'overall_recall': 0.6390346274921301, 'overall_f1': 0.6059701492537314, 'overall_accuracy': 0.826020619144719}
			------------EPOCH 27---------------
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0416, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1464, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0088, 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(0.0003, 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(0.0022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0249, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0684, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0861, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0296, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4580335731414868, 'recall': 0.5520231213872833, 'f1': 0.5006553079947575, 'number': 1038}, 'P': {'precision': 0.6360840594566889, 'recall': 0.6814936847885777, 'f1': 0.6580063626723224, 'number': 1821}, 'overall_precision': 0.5665209244222361, 'overall_recall': 0.6344875830710038, 'overall_f1': 0.5985810922290051, 'overall_accuracy': 0.8221848308117897}
			------------EPOCH 28---------------
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0379, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1681, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0212, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0176, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0148, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0129, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0618, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1273, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0499, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0150, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0246, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5063752276867031, 'recall': 0.535645472061657, 'f1': 0.5205992509363296, 'number': 1038}, 'P': {'precision': 0.6231032794909447, 'recall': 0.6990664470071389, 'f1': 0.6589026915113871, 'number': 1821}, 'overall_precision': 0.5822986310092327, 'overall_recall': 0.639734172787688, 'overall_f1': 0.6096666666666667, 'overall_accuracy': 0.8223844629183362}
			------------EPOCH 29---------------
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0296, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0039, 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(0.2067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0135, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0224, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0147, 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(0.0023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0236, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1180, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0846, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0457, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0249, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4675740592473979, 'recall': 0.5626204238921002, 'f1': 0.5107127240926979, 'number': 1038}, 'P': {'precision': 0.6281790437436419, 'recall': 0.6781987918725975, 'f1': 0.6522313176656984, 'number': 1821}, 'overall_precision': 0.5657853810264386, 'overall_recall': 0.6362364463098986, 'overall_f1': 0.5989463286137636, 'overall_accuracy': 0.821471859002695}
			------------EPOCH 30---------------
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0425, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0282, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0269, 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(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0202, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0927, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0629, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0246, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4977416440831075, 'recall': 0.5308285163776493, 'f1': 0.5137529137529138, 'number': 1038}, 'P': {'precision': 0.5829493087557603, 'recall': 0.6946732564524987, 'f1': 0.6339263342520671, 'number': 1821}, 'overall_precision': 0.5541653951785169, 'overall_recall': 0.6351871283665618, 'overall_f1': 0.5919165580182529, 'overall_accuracy': 0.8258637653467182}
