Precision: [tensor(0.8518, device='cuda:0'), tensor(0.8559, device='cuda:0'), tensor(0.8541, device='cuda:0'), tensor(0.8540, device='cuda:0'), tensor(0.8534, device='cuda:0'), tensor(0.8527, device='cuda:0'), tensor(0.8542, device='cuda:0'), tensor(0.8536, device='cuda:0'), tensor(0.8493, device='cuda:0'), tensor(0.8552, device='cuda:0')]

Output distance: [tensor(571.5781, device='cuda:0'), tensor(545.0524, device='cuda:0'), tensor(555.6028, device='cuda:0'), tensor(558.9155, device='cuda:0'), tensor(560.4969, device='cuda:0'), tensor(561.8216, device='cuda:0'), tensor(552.3605, device='cuda:0'), tensor(562.8864, device='cuda:0'), tensor(584.4297, device='cuda:0'), tensor(558.8113, device='cuda:0')]

Prediction loss: [tensor(596.1044, device='cuda:0'), tensor(582.2558, device='cuda:0'), tensor(621.7384, device='cuda:0'), tensor(598.3325, device='cuda:0'), tensor(599.3585, device='cuda:0'), tensor(592.7899, device='cuda:0'), tensor(613.1973, device='cuda:0'), tensor(606.8539, device='cuda:0'), tensor(587.8521, device='cuda:0'), tensor(584.3328, device='cuda:0')]

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

Compressed training loss: [tensor(8750298., device='cuda:0'), tensor(8536301., device='cuda:0'), tensor(9018398., device='cuda:0'), tensor(8800031., device='cuda:0'), tensor(8762318., device='cuda:0'), tensor(8727974., device='cuda:0'), tensor(8925490., device='cuda:0'), tensor(8898119., device='cuda:0'), tensor(8686218., device='cuda:0'), tensor(8672064., device='cuda:0')]

Training loss: 8823796.0

Prediction time: [datetime.timedelta(microseconds=759774), datetime.timedelta(microseconds=823507), datetime.timedelta(microseconds=805584), datetime.timedelta(microseconds=711982), datetime.timedelta(microseconds=792591), datetime.timedelta(microseconds=804589), datetime.timedelta(microseconds=863342), datetime.timedelta(microseconds=805535), datetime.timedelta(microseconds=806581), datetime.timedelta(microseconds=799609)]

Phi time: [datetime.timedelta(seconds=1, microseconds=299242), datetime.timedelta(microseconds=782145), datetime.timedelta(microseconds=707468), datetime.timedelta(microseconds=709241), datetime.timedelta(microseconds=716975), datetime.timedelta(microseconds=712839), datetime.timedelta(microseconds=698506), datetime.timedelta(microseconds=710385), datetime.timedelta(microseconds=702612), datetime.timedelta(microseconds=716514)]

