Precision: [tensor(0.9993, device='cuda:0'), tensor(0.9995, device='cuda:0'), tensor(0.9995, device='cuda:0'), tensor(0.9992, device='cuda:0'), tensor(0.9995, device='cuda:0'), tensor(0.9997, device='cuda:0'), tensor(0.9992, device='cuda:0'), tensor(0.9995, device='cuda:0'), tensor(0.9997, device='cuda:0'), tensor(0.9990, device='cuda:0')]

Output distance: [tensor(138986.2812, device='cuda:0'), tensor(139062.9375, device='cuda:0'), tensor(139157.5312, device='cuda:0'), tensor(138909.7969, device='cuda:0'), tensor(139088.5156, device='cuda:0'), tensor(139070.3438, device='cuda:0'), tensor(139903.5781, device='cuda:0'), tensor(138814.1562, device='cuda:0'), tensor(139021.0625, device='cuda:0'), tensor(139172.4219, device='cuda:0')]

Prediction loss: [tensor(136367.5156, device='cuda:0'), tensor(135676.2344, device='cuda:0'), tensor(132238.8125, device='cuda:0'), tensor(132817.5781, device='cuda:0'), tensor(134995.5469, device='cuda:0'), tensor(134160.0625, device='cuda:0'), tensor(138751.2344, device='cuda:0'), tensor(133563.5781, device='cuda:0'), tensor(131758.4062, device='cuda:0'), tensor(146435.8594, device='cuda:0')]

Others: [{'iter_num': 9, 'num_positive': tensor(6000, device='cuda:0'), 'num_positive_true': tensor(18000, device='cuda:0')}, {'iter_num': 11, 'num_positive': tensor(6000, device='cuda:0'), 'num_positive_true': tensor(18000, device='cuda:0')}, {'iter_num': 9, 'num_positive': tensor(6000, device='cuda:0'), 'num_positive_true': tensor(18000, device='cuda:0')}, {'iter_num': 11, 'num_positive': tensor(6000, device='cuda:0'), 'num_positive_true': tensor(18000, device='cuda:0')}, {'iter_num': 11, 'num_positive': tensor(6000, device='cuda:0'), 'num_positive_true': tensor(18000, device='cuda:0')}, {'iter_num': 11, 'num_positive': tensor(6000, device='cuda:0'), 'num_positive_true': tensor(18000, device='cuda:0')}, {'iter_num': 11, 'num_positive': tensor(6000, device='cuda:0'), 'num_positive_true': tensor(18000, device='cuda:0')}, {'iter_num': 9, 'num_positive': tensor(6000, device='cuda:0'), 'num_positive_true': tensor(18000, device='cuda:0')}, {'iter_num': 9, 'num_positive': tensor(6000, device='cuda:0'), 'num_positive_true': tensor(18000, device='cuda:0')}, {'iter_num': 9, 'num_positive': tensor(6000, device='cuda:0'), 'num_positive_true': tensor(18000, device='cuda:0')}]

Compressed training loss: [tensor(1.9089e+08, device='cuda:0'), tensor(1.9329e+08, device='cuda:0'), tensor(1.8862e+08, device='cuda:0'), tensor(1.9142e+08, device='cuda:0'), tensor(1.9304e+08, device='cuda:0'), tensor(1.9129e+08, device='cuda:0'), tensor(1.9179e+08, device='cuda:0'), tensor(1.8971e+08, device='cuda:0'), tensor(1.9037e+08, device='cuda:0'), tensor(1.9750e+08, device='cuda:0')]

Training loss: 191864736.0

Prediction time: [datetime.timedelta(microseconds=650242), datetime.timedelta(microseconds=756788), datetime.timedelta(microseconds=648250), datetime.timedelta(microseconds=737871), datetime.timedelta(microseconds=746832), datetime.timedelta(microseconds=729905), datetime.timedelta(microseconds=733888), datetime.timedelta(microseconds=653230), datetime.timedelta(microseconds=656217), datetime.timedelta(microseconds=653229)]

Phi time: [datetime.timedelta(seconds=1, microseconds=397068), datetime.timedelta(microseconds=902240), datetime.timedelta(microseconds=885844), datetime.timedelta(microseconds=860252), datetime.timedelta(microseconds=852959), datetime.timedelta(microseconds=860689), datetime.timedelta(microseconds=858468), datetime.timedelta(microseconds=877275), datetime.timedelta(microseconds=857303), datetime.timedelta(microseconds=859928)]

