Precision: [tensor(0.0730, device='cuda:0'), tensor(0.0487, device='cuda:0'), tensor(0.0883, device='cuda:0'), tensor(0.0661, device='cuda:0'), tensor(0.0845, device='cuda:0'), tensor(0.0558, device='cuda:0'), tensor(0.0738, device='cuda:0'), tensor(0.0874, device='cuda:0'), tensor(0.0603, device='cuda:0'), tensor(0.0896, device='cuda:0')]

Output distance: [tensor(19.8794, device='cuda:0'), tensor(19.9281, device='cuda:0'), tensor(19.8489, device='cuda:0'), tensor(19.8933, device='cuda:0'), tensor(19.8564, device='cuda:0'), tensor(19.9138, device='cuda:0'), tensor(19.8779, device='cuda:0'), tensor(19.8507, device='cuda:0'), tensor(19.9048, device='cuda:0'), tensor(19.8461, device='cuda:0')]

Prediction loss: [tensor(110.6914, device='cuda:0'), tensor(111.6111, device='cuda:0'), tensor(111.8580, device='cuda:0'), tensor(110.1722, device='cuda:0'), tensor(109.9624, device='cuda:0'), tensor(107.8676, device='cuda:0'), tensor(109.4786, device='cuda:0'), tensor(109.8277, device='cuda:0'), tensor(110.0225, device='cuda:0'), tensor(111.1884, device='cuda:0')]

Others: [{'iter_num': 30, 'num_positive': tensor(6616, device='cuda:0'), 'num_positive_true': tensor(125872, device='cuda:0')}, {'iter_num': 30, 'num_positive': tensor(6616, device='cuda:0'), 'num_positive_true': tensor(125872, device='cuda:0')}, {'iter_num': 30, 'num_positive': tensor(6616, device='cuda:0'), 'num_positive_true': tensor(125872, device='cuda:0')}, {'iter_num': 30, 'num_positive': tensor(6616, device='cuda:0'), 'num_positive_true': tensor(125872, device='cuda:0')}, {'iter_num': 30, 'num_positive': tensor(6616, device='cuda:0'), 'num_positive_true': tensor(125872, device='cuda:0')}, {'iter_num': 30, 'num_positive': tensor(6616, device='cuda:0'), 'num_positive_true': tensor(125872, device='cuda:0')}, {'iter_num': 30, 'num_positive': tensor(6616, device='cuda:0'), 'num_positive_true': tensor(125872, device='cuda:0')}, {'iter_num': 30, 'num_positive': tensor(6616, device='cuda:0'), 'num_positive_true': tensor(125872, device='cuda:0')}, {'iter_num': 30, 'num_positive': tensor(6616, device='cuda:0'), 'num_positive_true': tensor(125872, device='cuda:0')}, {'iter_num': 30, 'num_positive': tensor(6616, device='cuda:0'), 'num_positive_true': tensor(125872, device='cuda:0')}]

Compressed training loss: [tensor(0.0001, device='cuda:0'), tensor(0.0001, device='cuda:0'), tensor(0.0001, device='cuda:0'), tensor(0.0001, device='cuda:0'), tensor(0.0001, device='cuda:0'), tensor(0.0001, device='cuda:0'), tensor(0.0001, device='cuda:0'), tensor(0.0001, device='cuda:0'), tensor(0.0001, device='cuda:0'), tensor(0.0001, device='cuda:0')]

Training loss: 0

Prediction time: [datetime.timedelta(seconds=3, microseconds=347958), datetime.timedelta(seconds=3, microseconds=371432), datetime.timedelta(seconds=3, microseconds=344185), datetime.timedelta(seconds=3, microseconds=326537), datetime.timedelta(seconds=3, microseconds=353942), datetime.timedelta(seconds=3, microseconds=343181), datetime.timedelta(seconds=3, microseconds=345660), datetime.timedelta(seconds=3, microseconds=359221), datetime.timedelta(seconds=3, microseconds=370250), datetime.timedelta(seconds=3, microseconds=347937)]

Phi time: [datetime.timedelta(seconds=170, microseconds=305285), datetime.timedelta(seconds=170, microseconds=631040), datetime.timedelta(seconds=170, microseconds=394549), datetime.timedelta(seconds=170, microseconds=344146), datetime.timedelta(seconds=170, microseconds=269780), datetime.timedelta(seconds=170, microseconds=280116), datetime.timedelta(seconds=170, microseconds=515868), datetime.timedelta(seconds=170, microseconds=455091), datetime.timedelta(seconds=170, microseconds=199023), datetime.timedelta(seconds=170, microseconds=571669)]

