Precision: [tensor(0.4602, device='cuda:0'), tensor(0.4523, device='cuda:0'), tensor(0.4537, device='cuda:0'), tensor(0.4487, device='cuda:0'), tensor(0.4571, device='cuda:0'), tensor(0.4553, device='cuda:0'), tensor(0.4543, device='cuda:0'), tensor(0.4617, device='cuda:0'), tensor(0.4556, device='cuda:0'), tensor(0.4555, device='cuda:0')]
Output distance: [tensor(19.2279, device='cuda:0'), tensor(19.2680, device='cuda:0'), tensor(19.2610, device='cuda:0'), tensor(19.2858, device='cuda:0'), tensor(19.2435, device='cuda:0'), tensor(19.2529, device='cuda:0'), tensor(19.2576, device='cuda:0'), tensor(19.2204, device='cuda:0'), tensor(19.2511, device='cuda:0'), tensor(19.2524, device='cuda:0')]
Prediction loss: [tensor(104.4254, device='cuda:0'), tensor(103.9771, device='cuda:0'), tensor(104.9743, device='cuda:0'), tensor(103.7067, device='cuda:0'), tensor(104.5917, device='cuda:0'), tensor(104.3390, device='cuda:0'), tensor(104.7579, device='cuda:0'), tensor(104.3229, device='cuda:0'), tensor(104.8827, device='cuda:0'), tensor(104.5351, device='cuda:0')]
Others: [{'iter_num': 9, 'num_positive': tensor(16854, device='cuda:0'), 'num_positive_true': tensor(125872, device='cuda:0')}, {'iter_num': 9, 'num_positive': tensor(16827, device='cuda:0'), 'num_positive_true': tensor(125872, device='cuda:0')}, {'iter_num': 9, 'num_positive': tensor(16831, device='cuda:0'), 'num_positive_true': tensor(125872, device='cuda:0')}, {'iter_num': 9, 'num_positive': tensor(16789, device='cuda:0'), 'num_positive_true': tensor(125872, device='cuda:0')}, {'iter_num': 7, 'num_positive': tensor(16829, device='cuda:0'), 'num_positive_true': tensor(125872, device='cuda:0')}, {'iter_num': 9, 'num_positive': tensor(16845, device='cuda:0'), 'num_positive_true': tensor(125872, device='cuda:0')}, {'iter_num': 9, 'num_positive': tensor(16802, device='cuda:0'), 'num_positive_true': tensor(125872, device='cuda:0')}, {'iter_num': 9, 'num_positive': tensor(16826, device='cuda:0'), 'num_positive_true': tensor(125872, device='cuda:0')}, {'iter_num': 9, 'num_positive': tensor(16829, device='cuda:0'), 'num_positive_true': tensor(125872, device='cuda:0')}, {'iter_num': 9, 'num_positive': tensor(16886, 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=883447), datetime.timedelta(seconds=2, microseconds=867286), datetime.timedelta(seconds=2, microseconds=883371), datetime.timedelta(seconds=2, microseconds=866375), datetime.timedelta(seconds=2, microseconds=733330), datetime.timedelta(seconds=2, microseconds=883377), datetime.timedelta(seconds=2, microseconds=883594), datetime.timedelta(seconds=2, microseconds=866770), datetime.timedelta(seconds=2, microseconds=882882), datetime.timedelta(seconds=2, microseconds=866945)]
Phi time: [datetime.timedelta(seconds=99, microseconds=469311), datetime.timedelta(seconds=99, microseconds=601640), datetime.timedelta(seconds=99, microseconds=432079), datetime.timedelta(seconds=99, microseconds=399415), datetime.timedelta(seconds=99, microseconds=348569), datetime.timedelta(seconds=99, microseconds=249582), datetime.timedelta(seconds=99, microseconds=415442), datetime.timedelta(seconds=99, microseconds=432650), datetime.timedelta(seconds=99, microseconds=367356), datetime.timedelta(seconds=99, microseconds=451606)]
