Precision: [tensor(0.9998, device='cuda:0'), tensor(0.9995, device='cuda:0'), tensor(0.9998, device='cuda:0'), tensor(0.9998, device='cuda:0'), tensor(0.9998, device='cuda:0'), tensor(0.9998, device='cuda:0'), tensor(0.9997, device='cuda:0'), tensor(0.9998, device='cuda:0'), tensor(0.9998, device='cuda:0'), tensor(0.9997, device='cuda:0')]

Output distance: [tensor(148576.5938, device='cuda:0'), tensor(143774.5312, device='cuda:0'), tensor(144558.1875, device='cuda:0'), tensor(157663.7031, device='cuda:0'), tensor(143077.0469, device='cuda:0'), tensor(146602.3906, device='cuda:0'), tensor(142137.5312, device='cuda:0'), tensor(144607.1094, device='cuda:0'), tensor(145778.0469, device='cuda:0'), tensor(146211.3125, device='cuda:0')]

Prediction loss: [tensor(144733.9844, device='cuda:0'), tensor(141409.7656, device='cuda:0'), tensor(140153.0156, device='cuda:0'), tensor(158376.7656, device='cuda:0'), tensor(140695.6406, device='cuda:0'), tensor(144509.0781, device='cuda:0'), tensor(135191.0156, device='cuda:0'), tensor(134869.1250, device='cuda:0'), tensor(142900.4844, device='cuda:0'), tensor(141428.9844, device='cuda:0')]

Others: [{'iter_num': 29, 'num_positive': tensor(6000, device='cuda:0'), 'num_positive_true': tensor(18000, device='cuda:0')}, {'iter_num': 19, 'num_positive': tensor(6000, device='cuda:0'), 'num_positive_true': tensor(18000, device='cuda:0')}, {'iter_num': 23, 'num_positive': tensor(6000, device='cuda:0'), 'num_positive_true': tensor(18000, device='cuda:0')}, {'iter_num': 30, 'num_positive': tensor(6000, device='cuda:0'), 'num_positive_true': tensor(18000, device='cuda:0')}, {'iter_num': 19, 'num_positive': tensor(6000, device='cuda:0'), 'num_positive_true': tensor(18000, device='cuda:0')}, {'iter_num': 25, 'num_positive': tensor(6000, device='cuda:0'), 'num_positive_true': tensor(18000, device='cuda:0')}, {'iter_num': 11, 'num_positive': tensor(6000, device='cuda:0'), 'num_positive_true': tensor(18000, device='cuda:0')}, {'iter_num': 23, 'num_positive': tensor(6000, device='cuda:0'), 'num_positive_true': tensor(18000, device='cuda:0')}, {'iter_num': 25, 'num_positive': tensor(6000, device='cuda:0'), 'num_positive_true': tensor(18000, device='cuda:0')}, {'iter_num': 27, 'num_positive': tensor(6000, device='cuda:0'), 'num_positive_true': tensor(18000, device='cuda:0')}]

Compressed training loss: [tensor(1.9137e+08, device='cuda:0'), tensor(1.9048e+08, device='cuda:0'), tensor(1.9141e+08, device='cuda:0'), tensor(1.9435e+08, device='cuda:0'), tensor(1.9261e+08, device='cuda:0'), tensor(1.9341e+08, device='cuda:0'), tensor(1.9061e+08, device='cuda:0'), tensor(1.8712e+08, device='cuda:0'), tensor(1.9240e+08, device='cuda:0'), tensor(1.8876e+08, device='cuda:0')]

Training loss: 191708544.0

Prediction time: [datetime.timedelta(seconds=1, microseconds=829303), datetime.timedelta(seconds=1, microseconds=353303), datetime.timedelta(seconds=1, microseconds=516616), datetime.timedelta(seconds=1, microseconds=864152), datetime.timedelta(seconds=1, microseconds=315465), datetime.timedelta(seconds=1, microseconds=622169), datetime.timedelta(microseconds=887267), datetime.timedelta(seconds=1, microseconds=510640), datetime.timedelta(seconds=1, microseconds=633125), datetime.timedelta(seconds=1, microseconds=724740)]

Phi time: [datetime.timedelta(seconds=1, microseconds=538718), datetime.timedelta(seconds=1, microseconds=3720), datetime.timedelta(microseconds=934307), datetime.timedelta(microseconds=942516), datetime.timedelta(microseconds=941137), datetime.timedelta(microseconds=938308), datetime.timedelta(microseconds=940388), datetime.timedelta(microseconds=939334), datetime.timedelta(microseconds=949571), datetime.timedelta(microseconds=935874)]

