Precision: [tensor(0.8258, device='cuda:0'), tensor(0.8252, device='cuda:0'), tensor(0.8258, device='cuda:0'), tensor(0.8263, device='cuda:0'), tensor(0.8254, device='cuda:0'), tensor(0.8252, device='cuda:0'), tensor(0.8266, device='cuda:0'), tensor(0.8252, device='cuda:0'), tensor(0.8268, device='cuda:0'), tensor(0.8258, device='cuda:0')]

Output distance: [tensor(13988.7695, device='cuda:0'), tensor(13897.2568, device='cuda:0'), tensor(13794.2852, device='cuda:0'), tensor(13704.5107, device='cuda:0'), tensor(14194.6494, device='cuda:0'), tensor(13757.1475, device='cuda:0'), tensor(13653.7959, device='cuda:0'), tensor(13856.2559, device='cuda:0'), tensor(13676.6426, device='cuda:0'), tensor(13832.7344, device='cuda:0')]

Prediction loss: [tensor(10789.4131, device='cuda:0'), tensor(10633.2471, device='cuda:0'), tensor(10503.2217, device='cuda:0'), tensor(10360.6719, device='cuda:0'), tensor(10801.8408, device='cuda:0'), tensor(10327.4678, device='cuda:0'), tensor(10513.4121, device='cuda:0'), tensor(10543.9648, device='cuda:0'), tensor(10338.2051, device='cuda:0'), tensor(10631.0889, 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(1.9504e+08, device='cuda:0'), tensor(1.9359e+08, device='cuda:0'), tensor(1.9170e+08, device='cuda:0'), tensor(1.8994e+08, device='cuda:0'), tensor(1.9394e+08, device='cuda:0'), tensor(1.8968e+08, device='cuda:0'), tensor(1.9233e+08, device='cuda:0'), tensor(1.9241e+08, device='cuda:0'), tensor(1.8986e+08, device='cuda:0'), tensor(1.9369e+08, device='cuda:0')]

Training loss: 191970336.0

Prediction time: [datetime.timedelta(seconds=1, microseconds=748584), datetime.timedelta(seconds=1, microseconds=788416), datetime.timedelta(seconds=1, microseconds=765512), datetime.timedelta(seconds=1, microseconds=780449), datetime.timedelta(seconds=1, microseconds=813312), datetime.timedelta(seconds=1, microseconds=763521), datetime.timedelta(seconds=1, microseconds=787419), datetime.timedelta(seconds=1, microseconds=769493), datetime.timedelta(seconds=1, microseconds=789411), datetime.timedelta(seconds=1, microseconds=765512)]

Phi time: [datetime.timedelta(seconds=1, microseconds=380800), datetime.timedelta(microseconds=878883), datetime.timedelta(microseconds=845302), datetime.timedelta(microseconds=853579), datetime.timedelta(microseconds=851580), datetime.timedelta(microseconds=846911), datetime.timedelta(microseconds=846666), datetime.timedelta(microseconds=853038), datetime.timedelta(microseconds=852919), datetime.timedelta(microseconds=849260)]

