Precision: [tensor(0.4546, device='cuda:0'), tensor(0.4447, device='cuda:0'), tensor(0.4535, device='cuda:0'), tensor(0.4442, device='cuda:0'), tensor(0.4446, device='cuda:0'), tensor(0.4606, device='cuda:0'), tensor(0.4499, device='cuda:0'), tensor(0.4570, device='cuda:0'), tensor(0.4512, device='cuda:0'), tensor(0.4507, device='cuda:0')]
Output distance: [tensor(19.2562, device='cuda:0'), tensor(19.3058, device='cuda:0'), tensor(19.2619, device='cuda:0'), tensor(19.3096, device='cuda:0'), tensor(19.3068, device='cuda:0'), tensor(19.2267, device='cuda:0'), tensor(19.2801, device='cuda:0'), tensor(19.2441, device='cuda:0'), tensor(19.2733, device='cuda:0'), tensor(19.2760, device='cuda:0')]
Prediction loss: [tensor(104.4153, device='cuda:0'), tensor(104.3608, device='cuda:0'), tensor(104.5320, device='cuda:0'), tensor(104.7447, device='cuda:0'), tensor(104.0124, device='cuda:0'), tensor(105.8090, device='cuda:0'), tensor(104.3080, device='cuda:0'), tensor(103.9689, device='cuda:0'), tensor(104.5049, device='cuda:0'), tensor(104.9584, device='cuda:0')]
Others: [{'iter_num': 9, 'num_positive': tensor(16825, device='cuda:0'), 'num_positive_true': tensor(125872, device='cuda:0')}, {'iter_num': 9, 'num_positive': tensor(16785, device='cuda:0'), 'num_positive_true': tensor(125872, device='cuda:0')}, {'iter_num': 9, 'num_positive': tensor(16825, device='cuda:0'), 'num_positive_true': tensor(125872, device='cuda:0')}, {'iter_num': 7, 'num_positive': tensor(16850, device='cuda:0'), 'num_positive_true': tensor(125872, device='cuda:0')}, {'iter_num': 9, 'num_positive': tensor(16820, device='cuda:0'), 'num_positive_true': tensor(125872, device='cuda:0')}, {'iter_num': 7, 'num_positive': tensor(16888, device='cuda:0'), 'num_positive_true': tensor(125872, device='cuda:0')}, {'iter_num': 9, 'num_positive': tensor(16833, device='cuda:0'), 'num_positive_true': tensor(125872, device='cuda:0')}, {'iter_num': 7, 'num_positive': tensor(16839, device='cuda:0'), 'num_positive_true': tensor(125872, device='cuda:0')}, {'iter_num': 9, 'num_positive': tensor(16788, device='cuda:0'), 'num_positive_true': tensor(125872, device='cuda:0')}, {'iter_num': 9, 'num_positive': tensor(16828, 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=2, microseconds=899935), datetime.timedelta(seconds=2, microseconds=899919), datetime.timedelta(seconds=2, microseconds=900005), datetime.timedelta(seconds=2, microseconds=749910), datetime.timedelta(seconds=2, microseconds=899979), datetime.timedelta(seconds=2, microseconds=766695), datetime.timedelta(seconds=2, microseconds=899704), datetime.timedelta(seconds=2, microseconds=766291), datetime.timedelta(seconds=2, microseconds=916743), datetime.timedelta(seconds=2, microseconds=899998)]
Phi time: [datetime.timedelta(seconds=99, microseconds=287248), datetime.timedelta(seconds=99, microseconds=485986), datetime.timedelta(seconds=99, microseconds=416259), datetime.timedelta(seconds=99, microseconds=399420), datetime.timedelta(seconds=99, microseconds=399150), datetime.timedelta(seconds=99, microseconds=516220), datetime.timedelta(seconds=99, microseconds=466269), datetime.timedelta(seconds=99, microseconds=482687), datetime.timedelta(seconds=99, microseconds=666890), datetime.timedelta(seconds=99, microseconds=468298)]
