Precision: [tensor(0.8433, device='cuda:0'), tensor(0.8442, device='cuda:0'), tensor(0.8448, device='cuda:0'), tensor(0.8484, device='cuda:0'), tensor(0.8424, device='cuda:0'), tensor(0.8431, device='cuda:0'), tensor(0.8439, device='cuda:0'), tensor(0.8314, device='cuda:0'), tensor(0.8305, device='cuda:0'), tensor(0.8408, device='cuda:0')]

Output distance: [tensor(597.3427, device='cuda:0'), tensor(608.1645, device='cuda:0'), tensor(693.5536, device='cuda:0'), tensor(594.1863, device='cuda:0'), tensor(608.9728, device='cuda:0'), tensor(670.4675, device='cuda:0'), tensor(629.9780, device='cuda:0'), tensor(693.9699, device='cuda:0'), tensor(975.4555, device='cuda:0'), tensor(623.5031, device='cuda:0')]

Prediction loss: [tensor(600.9232, device='cuda:0'), tensor(614.4560, device='cuda:0'), tensor(774.7186, device='cuda:0'), tensor(559.9254, device='cuda:0'), tensor(571.7285, device='cuda:0'), tensor(676.3726, device='cuda:0'), tensor(633.1486, device='cuda:0'), tensor(592.8945, device='cuda:0'), tensor(1055.1456, device='cuda:0'), tensor(619.9913, device='cuda:0')]

Others: [{'iter_num': 30, 'num_positive': tensor(18000, device='cuda:0'), 'num_positive_true': tensor(18000, device='cuda:0')}, {'iter_num': 30, 'num_positive': tensor(18000, device='cuda:0'), 'num_positive_true': tensor(18000, device='cuda:0')}, {'iter_num': 30, 'num_positive': tensor(18000, device='cuda:0'), 'num_positive_true': tensor(18000, device='cuda:0')}, {'iter_num': 30, 'num_positive': tensor(18000, device='cuda:0'), 'num_positive_true': tensor(18000, device='cuda:0')}, {'iter_num': 30, 'num_positive': tensor(18000, device='cuda:0'), 'num_positive_true': tensor(18000, device='cuda:0')}, {'iter_num': 30, 'num_positive': tensor(18000, device='cuda:0'), 'num_positive_true': tensor(18000, device='cuda:0')}, {'iter_num': 30, 'num_positive': tensor(18000, device='cuda:0'), 'num_positive_true': tensor(18000, device='cuda:0')}, {'iter_num': 30, 'num_positive': tensor(18000, device='cuda:0'), 'num_positive_true': tensor(18000, device='cuda:0')}, {'iter_num': 30, 'num_positive': tensor(18000, device='cuda:0'), 'num_positive_true': tensor(18000, device='cuda:0')}, {'iter_num': 30, 'num_positive': tensor(18000, device='cuda:0'), 'num_positive_true': tensor(18000, device='cuda:0')}]

Compressed training loss: [tensor(8926268., device='cuda:0'), tensor(9180048., device='cuda:0'), tensor(9350567., device='cuda:0'), tensor(8517748., device='cuda:0'), tensor(8675389., device='cuda:0'), tensor(8726086., device='cuda:0'), tensor(9124386., device='cuda:0'), tensor(8383886.5000, device='cuda:0'), tensor(8938977., device='cuda:0'), tensor(9102920., device='cuda:0')]

Training loss: 8844968.0

Prediction time: [datetime.timedelta(seconds=1, microseconds=98370), datetime.timedelta(seconds=1, microseconds=148160), datetime.timedelta(seconds=1, microseconds=132228), datetime.timedelta(seconds=1, microseconds=122217), datetime.timedelta(seconds=1, microseconds=122271), datetime.timedelta(seconds=1, microseconds=116296), datetime.timedelta(seconds=1, microseconds=146168), datetime.timedelta(seconds=1, microseconds=131232), datetime.timedelta(seconds=1, microseconds=144178), datetime.timedelta(seconds=1, microseconds=162102)]

Phi time: [datetime.timedelta(seconds=1, microseconds=224514), datetime.timedelta(microseconds=711534), datetime.timedelta(microseconds=639234), datetime.timedelta(microseconds=643366), datetime.timedelta(microseconds=633755), datetime.timedelta(microseconds=632286), datetime.timedelta(microseconds=639601), datetime.timedelta(microseconds=637950), datetime.timedelta(microseconds=645436), datetime.timedelta(microseconds=634283)]

