Precision: [tensor(0.9992, device='cuda:0'), tensor(0.9993, device='cuda:0'), tensor(0.9992, device='cuda:0'), tensor(0.9988, device='cuda:0'), tensor(0.9990, device='cuda:0'), tensor(0.9993, device='cuda:0'), tensor(0.9992, device='cuda:0'), tensor(0.9988, device='cuda:0'), tensor(0.9992, device='cuda:0'), tensor(0.9988, device='cuda:0')]

Output distance: [tensor(140037.9219, device='cuda:0'), tensor(141533.3594, device='cuda:0'), tensor(154074.2656, device='cuda:0'), tensor(150190.3594, device='cuda:0'), tensor(144264.7500, device='cuda:0'), tensor(141443.3281, device='cuda:0'), tensor(142362., device='cuda:0'), tensor(143217.7188, device='cuda:0'), tensor(140232.6562, device='cuda:0'), tensor(145793.6875, device='cuda:0')]

Prediction loss: [tensor(136328.3906, device='cuda:0'), tensor(139668.0781, device='cuda:0'), tensor(149887.4219, device='cuda:0'), tensor(139356.2188, device='cuda:0'), tensor(145946.7812, device='cuda:0'), tensor(134523.9844, device='cuda:0'), tensor(137908.7812, device='cuda:0'), tensor(138552.4844, device='cuda:0'), tensor(133884.0469, device='cuda:0'), tensor(134378.9688, device='cuda:0')]

Others: [{'num_positive': tensor(6000, device='cuda:0'), 'num_positive_true': tensor(18000, device='cuda:0')}, {'num_positive': tensor(6000, device='cuda:0'), 'num_positive_true': tensor(18000, device='cuda:0')}, {'num_positive': tensor(6000, device='cuda:0'), 'num_positive_true': tensor(18000, device='cuda:0')}, {'num_positive': tensor(6000, device='cuda:0'), 'num_positive_true': tensor(18000, device='cuda:0')}, {'num_positive': tensor(6000, device='cuda:0'), 'num_positive_true': tensor(18000, device='cuda:0')}, {'num_positive': tensor(6000, device='cuda:0'), 'num_positive_true': tensor(18000, device='cuda:0')}, {'num_positive': tensor(6000, device='cuda:0'), 'num_positive_true': tensor(18000, device='cuda:0')}, {'num_positive': tensor(6000, device='cuda:0'), 'num_positive_true': tensor(18000, device='cuda:0')}, {'num_positive': tensor(6000, device='cuda:0'), 'num_positive_true': tensor(18000, device='cuda:0')}, {'num_positive': tensor(6000, device='cuda:0'), 'num_positive_true': tensor(18000, device='cuda:0')}]

Compressed training loss: [tensor(1.9238e+08, device='cuda:0'), tensor(1.9495e+08, device='cuda:0'), tensor(1.9707e+08, device='cuda:0'), tensor(1.9026e+08, device='cuda:0'), tensor(1.9783e+08, device='cuda:0'), tensor(1.9231e+08, device='cuda:0'), tensor(1.8994e+08, device='cuda:0'), tensor(1.9209e+08, device='cuda:0'), tensor(1.9215e+08, device='cuda:0'), tensor(1.9040e+08, device='cuda:0')]

Training loss: 191784976.0

Prediction time: [datetime.timedelta(microseconds=30866), datetime.timedelta(microseconds=30870), datetime.timedelta(microseconds=31866), datetime.timedelta(microseconds=31866), datetime.timedelta(microseconds=34851), datetime.timedelta(microseconds=32861), datetime.timedelta(microseconds=35849), datetime.timedelta(microseconds=34852), datetime.timedelta(microseconds=34852), datetime.timedelta(microseconds=30869)]

Phi time: [datetime.timedelta(seconds=1, microseconds=163753), datetime.timedelta(microseconds=633807), datetime.timedelta(microseconds=654044), datetime.timedelta(microseconds=638335), datetime.timedelta(microseconds=639348), datetime.timedelta(microseconds=633883), datetime.timedelta(microseconds=633377), datetime.timedelta(microseconds=636475), datetime.timedelta(microseconds=640444), datetime.timedelta(microseconds=635362)]

