Precision: [tensor(0.8120, device='cuda:0'), tensor(0.8163, device='cuda:0'), tensor(0.7608, device='cuda:0'), tensor(0.8189, device='cuda:0'), tensor(0.8129, device='cuda:0'), tensor(0.8009, device='cuda:0'), tensor(0.8092, device='cuda:0'), tensor(0.8206, device='cuda:0'), tensor(0.8170, device='cuda:0'), tensor(0.8142, device='cuda:0')]

Output distance: [tensor(15513.1514, device='cuda:0'), tensor(14396.8906, device='cuda:0'), tensor(41154.7148, device='cuda:0'), tensor(14223.8369, device='cuda:0'), tensor(15083.6855, device='cuda:0'), tensor(15698.0742, device='cuda:0'), tensor(15066.9121, device='cuda:0'), tensor(14235.3037, device='cuda:0'), tensor(14591.1123, device='cuda:0'), tensor(14683.1074, device='cuda:0')]

Prediction loss: [tensor(12080.7041, device='cuda:0'), tensor(9983.7549, device='cuda:0'), tensor(41688.8164, device='cuda:0'), tensor(10318.7998, device='cuda:0'), tensor(11076.6143, device='cuda:0'), tensor(10480.3066, device='cuda:0'), tensor(10906.7607, device='cuda:0'), tensor(10716.5088, device='cuda:0'), tensor(11302.3252, device='cuda:0'), tensor(10690.6289, 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(1.9569e+08, device='cuda:0'), tensor(1.8832e+08, device='cuda:0'), tensor(1.8474e+08, device='cuda:0'), tensor(1.9472e+08, device='cuda:0'), tensor(1.9345e+08, device='cuda:0'), tensor(1.8168e+08, device='cuda:0'), tensor(1.9566e+08, device='cuda:0'), tensor(1.9819e+08, device='cuda:0'), tensor(2.0322e+08, device='cuda:0'), tensor(1.9314e+08, device='cuda:0')]

Training loss: 192817952.0

Prediction time: [datetime.timedelta(microseconds=961921), datetime.timedelta(microseconds=986812), datetime.timedelta(microseconds=984824), datetime.timedelta(microseconds=973871), datetime.timedelta(microseconds=980840), datetime.timedelta(microseconds=972874), datetime.timedelta(microseconds=990795), datetime.timedelta(microseconds=981836), datetime.timedelta(microseconds=980841), datetime.timedelta(microseconds=976856)]

Phi time: [datetime.timedelta(seconds=1, microseconds=119186), datetime.timedelta(microseconds=656789), datetime.timedelta(microseconds=575455), datetime.timedelta(microseconds=581965), datetime.timedelta(microseconds=579227), datetime.timedelta(microseconds=579931), datetime.timedelta(microseconds=574322), datetime.timedelta(microseconds=582331), datetime.timedelta(microseconds=584242), datetime.timedelta(microseconds=585291)]

