Precision: [tensor(0.4355, device='cuda:0'), tensor(0.4261, device='cuda:0'), tensor(0.4317, device='cuda:0'), tensor(0.4296, device='cuda:0'), tensor(0.4317, device='cuda:0'), tensor(0.4323, device='cuda:0'), tensor(0.4285, device='cuda:0'), tensor(0.4325, device='cuda:0'), tensor(0.4279, device='cuda:0'), tensor(0.4351, device='cuda:0')]
Output distance: [tensor(19.4123, device='cuda:0'), tensor(19.4686, device='cuda:0'), tensor(19.4353, device='cuda:0'), tensor(19.4477, device='cuda:0'), tensor(19.4350, device='cuda:0'), tensor(19.4314, device='cuda:0'), tensor(19.4547, device='cuda:0'), tensor(19.4302, device='cuda:0'), tensor(19.4583, device='cuda:0'), tensor(19.4151, device='cuda:0')]
Prediction loss: [tensor(105.4212, device='cuda:0'), tensor(104.6315, device='cuda:0'), tensor(105.1734, device='cuda:0'), tensor(105.1641, device='cuda:0'), tensor(104.6831, device='cuda:0'), tensor(104.2206, device='cuda:0'), tensor(105.0374, device='cuda:0'), tensor(104.1859, device='cuda:0'), tensor(103.4992, device='cuda:0'), tensor(104.1964, device='cuda:0')]
Others: [{'iter_num': 7, 'num_positive': tensor(19848, device='cuda:0'), 'num_positive_true': tensor(125872, device='cuda:0')}, {'iter_num': 7, 'num_positive': tensor(19848, device='cuda:0'), 'num_positive_true': tensor(125872, device='cuda:0')}, {'iter_num': 9, 'num_positive': tensor(19848, device='cuda:0'), 'num_positive_true': tensor(125872, device='cuda:0')}, {'iter_num': 7, 'num_positive': tensor(19848, device='cuda:0'), 'num_positive_true': tensor(125872, device='cuda:0')}, {'iter_num': 7, 'num_positive': tensor(19848, device='cuda:0'), 'num_positive_true': tensor(125872, device='cuda:0')}, {'iter_num': 7, 'num_positive': tensor(19848, device='cuda:0'), 'num_positive_true': tensor(125872, device='cuda:0')}, {'iter_num': 7, 'num_positive': tensor(19848, device='cuda:0'), 'num_positive_true': tensor(125872, device='cuda:0')}, {'iter_num': 7, 'num_positive': tensor(19848, device='cuda:0'), 'num_positive_true': tensor(125872, device='cuda:0')}, {'iter_num': 7, 'num_positive': tensor(19848, device='cuda:0'), 'num_positive_true': tensor(125872, device='cuda:0')}, {'iter_num': 9, 'num_positive': tensor(19848, 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=733154), datetime.timedelta(seconds=2, microseconds=766775), datetime.timedelta(seconds=2, microseconds=899966), datetime.timedelta(seconds=2, microseconds=750103), datetime.timedelta(seconds=2, microseconds=766383), datetime.timedelta(seconds=2, microseconds=766498), datetime.timedelta(seconds=2, microseconds=750550), datetime.timedelta(seconds=2, microseconds=766566), datetime.timedelta(seconds=2, microseconds=749885), datetime.timedelta(seconds=2, microseconds=917827)]
Phi time: [datetime.timedelta(seconds=99, microseconds=513634), datetime.timedelta(seconds=99, microseconds=451262), datetime.timedelta(seconds=99, microseconds=399550), datetime.timedelta(seconds=99, microseconds=416075), datetime.timedelta(seconds=99, microseconds=416013), datetime.timedelta(seconds=99, microseconds=482762), datetime.timedelta(seconds=99, microseconds=416174), datetime.timedelta(seconds=99, microseconds=781885), datetime.timedelta(seconds=99, microseconds=450232), datetime.timedelta(seconds=99, microseconds=451697)]
