Precision: [tensor(0.8274, device='cuda:0'), tensor(0.8266, device='cuda:0'), tensor(0.8266, device='cuda:0'), tensor(0.8263, device='cuda:0'), tensor(0.8266, device='cuda:0'), tensor(0.8267, device='cuda:0'), tensor(0.8278, device='cuda:0'), tensor(0.8265, device='cuda:0'), tensor(0.8251, device='cuda:0'), tensor(0.8260, device='cuda:0')]

Output distance: [tensor(13554.9688, device='cuda:0'), tensor(13607.7656, device='cuda:0'), tensor(13622.9062, device='cuda:0'), tensor(13649.4023, device='cuda:0'), tensor(13613.3516, device='cuda:0'), tensor(13589.2002, device='cuda:0'), tensor(13554.5264, device='cuda:0'), tensor(13615.6680, device='cuda:0'), tensor(13724.4521, device='cuda:0'), tensor(13629.6025, device='cuda:0')]

Prediction loss: [tensor(10503.3486, device='cuda:0'), tensor(10615.2383, device='cuda:0'), tensor(10476.1768, device='cuda:0'), tensor(10574.5371, device='cuda:0'), tensor(10566.0684, device='cuda:0'), tensor(10642.2803, device='cuda:0'), tensor(10704.1074, device='cuda:0'), tensor(10469.8516, device='cuda:0'), tensor(10673.2363, device='cuda:0'), tensor(10454.2803, device='cuda:0')]

Others: [{'iter_num': 13, 'num_positive': tensor(18000, device='cuda:0'), 'num_positive_true': tensor(18000, device='cuda:0')}, {'iter_num': 11, 'num_positive': tensor(18000, device='cuda:0'), 'num_positive_true': tensor(18000, device='cuda:0')}, {'iter_num': 11, 'num_positive': tensor(18000, device='cuda:0'), 'num_positive_true': tensor(18000, device='cuda:0')}, {'iter_num': 13, 'num_positive': tensor(18000, device='cuda:0'), 'num_positive_true': tensor(18000, device='cuda:0')}, {'iter_num': 11, 'num_positive': tensor(18000, device='cuda:0'), 'num_positive_true': tensor(18000, device='cuda:0')}, {'iter_num': 11, 'num_positive': tensor(18000, device='cuda:0'), 'num_positive_true': tensor(18000, device='cuda:0')}, {'iter_num': 13, 'num_positive': tensor(18000, device='cuda:0'), 'num_positive_true': tensor(18000, device='cuda:0')}, {'iter_num': 13, 'num_positive': tensor(18000, device='cuda:0'), 'num_positive_true': tensor(18000, device='cuda:0')}, {'iter_num': 11, 'num_positive': tensor(18000, device='cuda:0'), 'num_positive_true': tensor(18000, device='cuda:0')}, {'iter_num': 13, 'num_positive': tensor(18000, device='cuda:0'), 'num_positive_true': tensor(18000, device='cuda:0')}]

Compressed training loss: [tensor(1.9101e+08, device='cuda:0'), tensor(1.9344e+08, device='cuda:0'), tensor(1.9079e+08, device='cuda:0'), tensor(1.9254e+08, device='cuda:0'), tensor(1.9187e+08, device='cuda:0'), tensor(1.9316e+08, device='cuda:0'), tensor(1.9473e+08, device='cuda:0'), tensor(1.9051e+08, device='cuda:0'), tensor(1.9444e+08, device='cuda:0'), tensor(1.9038e+08, device='cuda:0')]

Training loss: 192092288.0

Prediction time: [datetime.timedelta(microseconds=848429), datetime.timedelta(microseconds=797645), datetime.timedelta(microseconds=784699), datetime.timedelta(microseconds=853411), datetime.timedelta(microseconds=784700), datetime.timedelta(microseconds=782708), datetime.timedelta(microseconds=867352), datetime.timedelta(microseconds=862374), datetime.timedelta(microseconds=785696), datetime.timedelta(microseconds=856397)]

Phi time: [datetime.timedelta(seconds=1, microseconds=500851), datetime.timedelta(microseconds=922079), datetime.timedelta(microseconds=871204), datetime.timedelta(microseconds=867811), datetime.timedelta(microseconds=865366), datetime.timedelta(microseconds=868697), datetime.timedelta(microseconds=866893), datetime.timedelta(microseconds=871158), datetime.timedelta(microseconds=866273), datetime.timedelta(microseconds=866243)]

