Precision: [tensor(0.9584, device='cuda:0'), tensor(0.9546, device='cuda:0'), tensor(0.9554, device='cuda:0'), tensor(0.9568, device='cuda:0'), tensor(0.9533, device='cuda:0'), tensor(0.9577, device='cuda:0'), tensor(0.9526, device='cuda:0'), tensor(0.9526, device='cuda:0'), tensor(0.9550, device='cuda:0'), tensor(0.9562, device='cuda:0')]

Output distance: [tensor(110.0917, device='cuda:0'), tensor(382.7350, device='cuda:0'), tensor(133.8172, device='cuda:0'), tensor(118.1532, device='cuda:0'), tensor(131.4480, device='cuda:0'), tensor(143.8849, device='cuda:0'), tensor(141.5944, device='cuda:0'), tensor(159.0772, device='cuda:0'), tensor(802.4932, device='cuda:0'), tensor(113.7955, device='cuda:0')]

Prediction loss: [tensor(366.4546, device='cuda:0'), tensor(734.2948, device='cuda:0'), tensor(403.2028, device='cuda:0'), tensor(385.9308, device='cuda:0'), tensor(379.0424, device='cuda:0'), tensor(410.8834, device='cuda:0'), tensor(394.3958, device='cuda:0'), tensor(413.1985, device='cuda:0'), tensor(1317.2545, device='cuda:0'), tensor(366.9850, 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(3472030.2500, device='cuda:0'), tensor(3839171.5000, device='cuda:0'), tensor(3660967., device='cuda:0'), tensor(3670626.5000, device='cuda:0'), tensor(3543398.5000, device='cuda:0'), tensor(3598801.2500, device='cuda:0'), tensor(3654020.2500, device='cuda:0'), tensor(3726368.7500, device='cuda:0'), tensor(3618047.5000, device='cuda:0'), tensor(3523922.2500, device='cuda:0')]

Training loss: 3618015.5

Prediction time: [datetime.timedelta(seconds=1, microseconds=616141), datetime.timedelta(seconds=1, microseconds=676888), datetime.timedelta(seconds=1, microseconds=651993), datetime.timedelta(seconds=1, microseconds=684854), datetime.timedelta(seconds=1, microseconds=669917), datetime.timedelta(seconds=1, microseconds=660956), datetime.timedelta(seconds=1, microseconds=658986), datetime.timedelta(seconds=1, microseconds=668922), datetime.timedelta(seconds=1, microseconds=674897), datetime.timedelta(seconds=1, microseconds=668922)]

Phi time: [datetime.timedelta(seconds=1, microseconds=381147), datetime.timedelta(microseconds=863104), datetime.timedelta(microseconds=799853), datetime.timedelta(microseconds=798974), datetime.timedelta(microseconds=803005), datetime.timedelta(microseconds=796481), datetime.timedelta(microseconds=798835), datetime.timedelta(microseconds=798382), datetime.timedelta(microseconds=798826), datetime.timedelta(microseconds=800494)]

