Precision: [tensor(0.8547, device='cuda:0'), tensor(0.8528, device='cuda:0'), tensor(0.8533, device='cuda:0'), tensor(0.8531, device='cuda:0'), tensor(0.8519, device='cuda:0'), tensor(0.8521, device='cuda:0'), tensor(0.8533, device='cuda:0'), tensor(0.8526, device='cuda:0'), tensor(0.8537, device='cuda:0'), tensor(0.8532, device='cuda:0')]

Output distance: [tensor(543.6869, device='cuda:0'), tensor(559.0125, device='cuda:0'), tensor(552.5171, device='cuda:0'), tensor(552.9618, device='cuda:0'), tensor(562.5800, device='cuda:0'), tensor(557.6692, device='cuda:0'), tensor(551.0605, device='cuda:0'), tensor(554.9296, device='cuda:0'), tensor(544.7328, device='cuda:0'), tensor(549.5891, device='cuda:0')]

Prediction loss: [tensor(612.7294, device='cuda:0'), tensor(598.5310, device='cuda:0'), tensor(583.9489, device='cuda:0'), tensor(615.0330, device='cuda:0'), tensor(597.2827, device='cuda:0'), tensor(572.4073, device='cuda:0'), tensor(590.8851, device='cuda:0'), tensor(591.3763, device='cuda:0'), tensor(598.9304, device='cuda:0'), tensor(592.4988, device='cuda:0')]

Others: [{'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': 13, 'num_positive': tensor(18000, device='cuda:0'), 'num_positive_true': tensor(18000, device='cuda:0')}, {'iter_num': 11, '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': 13, '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': 13, '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')}]

Compressed training loss: [tensor(8995036., device='cuda:0'), tensor(8882169., device='cuda:0'), tensor(8595117., device='cuda:0'), tensor(9026923., device='cuda:0'), tensor(8814525., device='cuda:0'), tensor(8454770., device='cuda:0'), tensor(8761617., device='cuda:0'), tensor(8721025., device='cuda:0'), tensor(8837378., device='cuda:0'), tensor(8747868., device='cuda:0')]

Training loss: 8835396.0

Prediction time: [datetime.timedelta(microseconds=868350), datetime.timedelta(microseconds=881295), datetime.timedelta(microseconds=751842), datetime.timedelta(microseconds=706039), datetime.timedelta(microseconds=775741), datetime.timedelta(microseconds=879307), datetime.timedelta(microseconds=791674), datetime.timedelta(microseconds=788687), datetime.timedelta(microseconds=795658), datetime.timedelta(microseconds=809601)]

Phi time: [datetime.timedelta(microseconds=807437), datetime.timedelta(microseconds=776260), datetime.timedelta(microseconds=744259), datetime.timedelta(microseconds=766598), datetime.timedelta(microseconds=780320), datetime.timedelta(microseconds=776534), datetime.timedelta(microseconds=791247), datetime.timedelta(microseconds=775645), datetime.timedelta(microseconds=800799), datetime.timedelta(microseconds=778368)]

