Precision: [tensor(0.8601, device='cuda:0'), tensor(0.8584, device='cuda:0'), tensor(0.8577, device='cuda:0'), tensor(0.8580, device='cuda:0'), tensor(0.8589, device='cuda:0'), tensor(0.8583, device='cuda:0'), tensor(0.8591, device='cuda:0'), tensor(0.8593, device='cuda:0'), tensor(0.8601, device='cuda:0'), tensor(0.8587, device='cuda:0')]

Output distance: [tensor(525.8183, device='cuda:0'), tensor(530.4252, device='cuda:0'), tensor(541.9548, device='cuda:0'), tensor(538.6618, device='cuda:0'), tensor(532.8981, device='cuda:0'), tensor(537.6014, device='cuda:0'), tensor(534.3906, device='cuda:0'), tensor(528.8582, device='cuda:0'), tensor(527.1368, device='cuda:0'), tensor(531.6823, device='cuda:0')]

Prediction loss: [tensor(602.4000, device='cuda:0'), tensor(581.1859, device='cuda:0'), tensor(593.8665, device='cuda:0'), tensor(605.4012, device='cuda:0'), tensor(622.7674, device='cuda:0'), tensor(622.1650, device='cuda:0'), tensor(587.8666, device='cuda:0'), tensor(586.7413, device='cuda:0'), tensor(588.7591, device='cuda:0'), tensor(598.2313, device='cuda:0')]

Others: [{'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')}, {'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': 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(8946689., device='cuda:0'), tensor(8638370., device='cuda:0'), tensor(8868717., device='cuda:0'), tensor(8981653., device='cuda:0'), tensor(9257444., device='cuda:0'), tensor(9208464., device='cuda:0'), tensor(8766798., device='cuda:0'), tensor(8732368., device='cuda:0'), tensor(8803611., device='cuda:0'), tensor(8888146., device='cuda:0')]

Training loss: 8921521.0

Prediction time: [datetime.timedelta(microseconds=782683), datetime.timedelta(microseconds=807575), datetime.timedelta(microseconds=787657), datetime.timedelta(microseconds=766747), datetime.timedelta(microseconds=732890), datetime.timedelta(microseconds=713972), datetime.timedelta(microseconds=790647), datetime.timedelta(microseconds=817534), datetime.timedelta(microseconds=792640), datetime.timedelta(microseconds=801600)]

Phi time: [datetime.timedelta(seconds=1, microseconds=437173), datetime.timedelta(microseconds=866425), datetime.timedelta(microseconds=802330), datetime.timedelta(microseconds=799875), datetime.timedelta(microseconds=805562), datetime.timedelta(microseconds=800368), datetime.timedelta(microseconds=798371), datetime.timedelta(microseconds=799707), datetime.timedelta(microseconds=801334), datetime.timedelta(microseconds=801868)]

