Precision: [tensor(0.8562, device='cuda:0'), tensor(0.8543, device='cuda:0'), tensor(0.8559, device='cuda:0'), tensor(0.8558, device='cuda:0'), tensor(0.8569, device='cuda:0'), tensor(0.8558, device='cuda:0'), tensor(0.8561, device='cuda:0'), tensor(0.8551, device='cuda:0'), tensor(0.8557, device='cuda:0'), tensor(0.8513, device='cuda:0')]

Output distance: [tensor(542.0484, device='cuda:0'), tensor(556.6134, device='cuda:0'), tensor(549.5200, device='cuda:0'), tensor(548.3279, device='cuda:0'), tensor(541.4431, device='cuda:0'), tensor(548.6387, device='cuda:0'), tensor(542.0612, device='cuda:0'), tensor(554.3809, device='cuda:0'), tensor(545.9012, device='cuda:0'), tensor(571.1583, device='cuda:0')]

Prediction loss: [tensor(591.6993, device='cuda:0'), tensor(572.7250, device='cuda:0'), tensor(581.6523, device='cuda:0'), tensor(621.1682, device='cuda:0'), tensor(603.8850, device='cuda:0'), tensor(553.4626, device='cuda:0'), tensor(568.1065, device='cuda:0'), tensor(648.6768, device='cuda:0'), tensor(590.0198, device='cuda:0'), tensor(602.9347, device='cuda:0')]

Others: [{'iter_num': 17, 'num_positive': tensor(18000, device='cuda:0'), 'num_positive_true': tensor(18000, device='cuda:0')}, {'iter_num': 15, 'num_positive': tensor(18000, device='cuda:0'), 'num_positive_true': tensor(18000, device='cuda:0')}, {'iter_num': 15, 'num_positive': tensor(18000, device='cuda:0'), 'num_positive_true': tensor(18000, device='cuda:0')}, {'iter_num': 13, 'num_positive': tensor(18000, device='cuda:0'), 'num_positive_true': tensor(18000, device='cuda:0')}, {'iter_num': 15, 'num_positive': tensor(18000, device='cuda:0'), 'num_positive_true': tensor(18000, device='cuda:0')}, {'iter_num': 17, 'num_positive': tensor(18000, device='cuda:0'), 'num_positive_true': tensor(18000, device='cuda:0')}, {'iter_num': 17, 'num_positive': tensor(18000, device='cuda:0'), 'num_positive_true': tensor(18000, device='cuda:0')}, {'iter_num': 15, 'num_positive': tensor(18000, device='cuda:0'), 'num_positive_true': tensor(18000, device='cuda:0')}, {'iter_num': 15, 'num_positive': tensor(18000, device='cuda:0'), 'num_positive_true': tensor(18000, device='cuda:0')}, {'iter_num': 17, 'num_positive': tensor(18000, device='cuda:0'), 'num_positive_true': tensor(18000, device='cuda:0')}]

Compressed training loss: [tensor(8719756., device='cuda:0'), tensor(8452203., device='cuda:0'), tensor(8699418., device='cuda:0'), tensor(9287308., device='cuda:0'), tensor(8879878., device='cuda:0'), tensor(8265439.5000, device='cuda:0'), tensor(8539820., device='cuda:0'), tensor(9581696., device='cuda:0'), tensor(8771716., device='cuda:0'), tensor(8985742., device='cuda:0')]

Training loss: 8871296.0

Prediction time: [datetime.timedelta(microseconds=793635), datetime.timedelta(microseconds=766748), datetime.timedelta(microseconds=767744), datetime.timedelta(microseconds=684098), datetime.timedelta(microseconds=762811), datetime.timedelta(microseconds=807574), datetime.timedelta(microseconds=826495), datetime.timedelta(microseconds=757802), datetime.timedelta(microseconds=769735), datetime.timedelta(microseconds=807575)]

Phi time: [datetime.timedelta(seconds=1, microseconds=286693), datetime.timedelta(microseconds=770604), datetime.timedelta(microseconds=727222), datetime.timedelta(microseconds=705416), datetime.timedelta(microseconds=704028), datetime.timedelta(microseconds=705046), datetime.timedelta(microseconds=706185), datetime.timedelta(microseconds=704859), datetime.timedelta(microseconds=707107), datetime.timedelta(microseconds=711526)]

