Precision: [tensor(0.2678, device='cuda:0'), tensor(0.1993, device='cuda:0'), tensor(0.1504, device='cuda:0'), tensor(0.1201, device='cuda:0'), tensor(0.1997, device='cuda:0'), tensor(0.2156, device='cuda:0'), tensor(0.1586, device='cuda:0'), tensor(0.2143, device='cuda:0'), tensor(0.2188, device='cuda:0'), tensor(0.2511, device='cuda:0')]

Output distance: [tensor(2.5500e+24, device='cuda:0'), tensor(6.8505e+23, device='cuda:0'), tensor(3.1737e+24, device='cuda:0'), tensor(1.5935e+23, device='cuda:0'), tensor(1.8838e+24, device='cuda:0'), tensor(1.4483e+23, device='cuda:0'), tensor(2.2848e+23, device='cuda:0'), tensor(2.4256e+23, device='cuda:0'), tensor(1.5208e+23, device='cuda:0'), tensor(9.6385e+23, device='cuda:0')]

Prediction loss: [tensor(4.6208e+24, device='cuda:0'), tensor(1.1080e+24, device='cuda:0'), tensor(4.6547e+24, device='cuda:0'), tensor(2.6628e+23, device='cuda:0'), tensor(3.4574e+24, device='cuda:0'), tensor(2.6451e+23, device='cuda:0'), tensor(3.8516e+23, device='cuda:0'), tensor(4.3764e+23, device='cuda:0'), tensor(2.5760e+23, device='cuda:0'), tensor(1.6998e+24, 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(17986, 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(17985, 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(1.9469e+08, device='cuda:0'), tensor(1.9388e+08, device='cuda:0'), tensor(1.8406e+08, device='cuda:0'), tensor(1.9329e+08, device='cuda:0'), tensor(1.8762e+08, device='cuda:0'), tensor(1.9071e+08, device='cuda:0'), tensor(1.9449e+08, device='cuda:0'), tensor(1.9082e+08, device='cuda:0'), tensor(1.9243e+08, device='cuda:0'), tensor(1.8459e+08, device='cuda:0')]

Training loss: 191958128.0

Prediction time: [datetime.timedelta(seconds=1, microseconds=195929), datetime.timedelta(seconds=1, microseconds=216841), datetime.timedelta(seconds=1, microseconds=235760), datetime.timedelta(seconds=1, microseconds=238749), datetime.timedelta(seconds=1, microseconds=238746), datetime.timedelta(seconds=1, microseconds=219827), datetime.timedelta(seconds=1, microseconds=237754), datetime.timedelta(seconds=1, microseconds=238746), datetime.timedelta(seconds=1, microseconds=237752), datetime.timedelta(seconds=1, microseconds=244720)]

Phi time: [datetime.timedelta(seconds=1, microseconds=107337), datetime.timedelta(microseconds=656219), datetime.timedelta(microseconds=573524), datetime.timedelta(microseconds=576555), datetime.timedelta(microseconds=583524), datetime.timedelta(microseconds=577550), datetime.timedelta(microseconds=577213), datetime.timedelta(microseconds=576247), datetime.timedelta(microseconds=579716), datetime.timedelta(microseconds=577147)]

