Precision: [tensor(0.9606, device='cuda:0'), tensor(0.9609, device='cuda:0'), tensor(0.9634, device='cuda:0'), tensor(0.9629, device='cuda:0'), tensor(0.9593, device='cuda:0'), tensor(0.9589, device='cuda:0'), tensor(0.9598, device='cuda:0'), tensor(0.9600, device='cuda:0'), tensor(0.9603, device='cuda:0'), tensor(0.9620, device='cuda:0')]

Output distance: [tensor(107.2546, device='cuda:0'), tensor(104.0981, device='cuda:0'), tensor(97.5369, device='cuda:0'), tensor(98.7897, device='cuda:0'), tensor(110.0575, device='cuda:0'), tensor(112.2441, device='cuda:0'), tensor(111.0735, device='cuda:0'), tensor(105.3825, device='cuda:0'), tensor(106.8416, device='cuda:0'), tensor(102.8494, device='cuda:0')]

Prediction loss: [tensor(377.6576, device='cuda:0'), tensor(378.9492, device='cuda:0'), tensor(385.4682, device='cuda:0'), tensor(389.1932, device='cuda:0'), tensor(385.9693, device='cuda:0'), tensor(381.4235, device='cuda:0'), tensor(381.7415, device='cuda:0'), tensor(377.0470, device='cuda:0'), tensor(395.6402, device='cuda:0'), tensor(387.9576, device='cuda:0')]

Others: [{'iter_num': 11, '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': 11, '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': 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': 11, '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': 9, 'num_positive': tensor(18000, device='cuda:0'), 'num_positive_true': tensor(18000, device='cuda:0')}, {'iter_num': 9, 'num_positive': tensor(18000, device='cuda:0'), 'num_positive_true': tensor(18000, device='cuda:0')}]

Compressed training loss: [tensor(3513344., device='cuda:0'), tensor(3519765.2500, device='cuda:0'), tensor(3573183., device='cuda:0'), tensor(3611209.7500, device='cuda:0'), tensor(3580039.5000, device='cuda:0'), tensor(3543500.7500, device='cuda:0'), tensor(3552121.5000, device='cuda:0'), tensor(3498690.2500, device='cuda:0'), tensor(3675776.5000, device='cuda:0'), tensor(3594659.5000, device='cuda:0')]

Training loss: 3574339.25

Prediction time: [datetime.timedelta(seconds=1, microseconds=111287), datetime.timedelta(seconds=1, microseconds=138172), datetime.timedelta(seconds=1, microseconds=129213), datetime.timedelta(seconds=1, microseconds=129211), datetime.timedelta(seconds=1, microseconds=124232), datetime.timedelta(seconds=1, microseconds=262645), datetime.timedelta(seconds=1, microseconds=129211), datetime.timedelta(seconds=1, microseconds=121245), datetime.timedelta(microseconds=969884), datetime.timedelta(microseconds=977851)]

Phi time: [datetime.timedelta(seconds=1, microseconds=851400), datetime.timedelta(seconds=1, microseconds=254066), datetime.timedelta(seconds=1, microseconds=310557), datetime.timedelta(seconds=1, microseconds=291910), datetime.timedelta(seconds=1, microseconds=288976), datetime.timedelta(seconds=1, microseconds=291713), datetime.timedelta(seconds=1, microseconds=286797), datetime.timedelta(seconds=1, microseconds=288206), datetime.timedelta(seconds=1, microseconds=314864), datetime.timedelta(seconds=1, microseconds=304563)]

