Precision: [tensor(0.7785, device='cuda:0'), tensor(0.7570, device='cuda:0'), tensor(0.7487, device='cuda:0'), tensor(0.7484, device='cuda:0'), tensor(0.7592, device='cuda:0'), tensor(0.7914, device='cuda:0'), tensor(0.7538, device='cuda:0'), tensor(0.7712, device='cuda:0'), tensor(0.7686, device='cuda:0'), tensor(0.7684, device='cuda:0')]

Output distance: [tensor(41006.8672, device='cuda:0'), tensor(33501.6836, device='cuda:0'), tensor(35866.4688, device='cuda:0'), tensor(36615.5273, device='cuda:0'), tensor(37032.5547, device='cuda:0'), tensor(29139.2637, device='cuda:0'), tensor(35130.1719, device='cuda:0'), tensor(35456.9102, device='cuda:0'), tensor(32718.3203, device='cuda:0'), tensor(31935.3086, device='cuda:0')]

Prediction loss: [tensor(40657.9297, device='cuda:0'), tensor(33888.8711, device='cuda:0'), tensor(35605.4609, device='cuda:0'), tensor(34975.9219, device='cuda:0'), tensor(34416.2227, device='cuda:0'), tensor(26893.6816, device='cuda:0'), tensor(33933.3594, device='cuda:0'), tensor(32277.0547, device='cuda:0'), tensor(29913.8613, device='cuda:0'), tensor(31576.7227, device='cuda:0')]

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

Compressed training loss: [tensor(1.8764e+08, device='cuda:0'), tensor(1.9135e+08, device='cuda:0'), tensor(1.8946e+08, device='cuda:0'), tensor(1.9370e+08, device='cuda:0'), tensor(1.9131e+08, device='cuda:0'), tensor(1.9132e+08, device='cuda:0'), tensor(1.9027e+08, device='cuda:0'), tensor(1.9142e+08, device='cuda:0'), tensor(1.8774e+08, device='cuda:0'), tensor(1.9065e+08, device='cuda:0')]

Training loss: 191077488.0

Prediction time: [datetime.timedelta(microseconds=85640), datetime.timedelta(microseconds=91615), datetime.timedelta(microseconds=84646), datetime.timedelta(microseconds=82655), datetime.timedelta(microseconds=80663), datetime.timedelta(microseconds=84643), datetime.timedelta(microseconds=82653), datetime.timedelta(microseconds=80661), datetime.timedelta(microseconds=83651), datetime.timedelta(microseconds=83649)]

Phi time: [datetime.timedelta(seconds=1, microseconds=517621), datetime.timedelta(microseconds=898225), datetime.timedelta(microseconds=861437), datetime.timedelta(microseconds=912656), datetime.timedelta(microseconds=885287), datetime.timedelta(microseconds=882144), datetime.timedelta(microseconds=879309), datetime.timedelta(microseconds=879535), datetime.timedelta(microseconds=877283), datetime.timedelta(microseconds=883835)]

