Precision: [tensor(0.8277, device='cuda:0'), tensor(0.8289, device='cuda:0'), tensor(0.8287, device='cuda:0'), tensor(0.8281, device='cuda:0'), tensor(0.8288, device='cuda:0'), tensor(0.8276, device='cuda:0'), tensor(0.8286, device='cuda:0'), tensor(0.8283, device='cuda:0'), tensor(0.8287, device='cuda:0'), tensor(0.8282, device='cuda:0')]

Output distance: [tensor(13822.6943, device='cuda:0'), tensor(13625.0518, device='cuda:0'), tensor(13749.3408, device='cuda:0'), tensor(13762.6670, device='cuda:0'), tensor(13745.1367, device='cuda:0'), tensor(13970.9023, device='cuda:0'), tensor(13755.3994, device='cuda:0'), tensor(13731.2148, device='cuda:0'), tensor(13595.9561, device='cuda:0'), tensor(13777.3691, device='cuda:0')]

Prediction loss: [tensor(10725.9971, device='cuda:0'), tensor(10659.9785, device='cuda:0'), tensor(10800.2188, device='cuda:0'), tensor(10752.4062, device='cuda:0'), tensor(10687.3770, device='cuda:0'), tensor(10826.8008, device='cuda:0'), tensor(10472.4639, device='cuda:0'), tensor(10698.3252, device='cuda:0'), tensor(10465.8564, device='cuda:0'), tensor(10792.1230, 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(1.9272e+08, device='cuda:0'), tensor(1.9192e+08, device='cuda:0'), tensor(1.9347e+08, device='cuda:0'), tensor(1.9248e+08, device='cuda:0'), tensor(1.9115e+08, device='cuda:0'), tensor(1.9259e+08, device='cuda:0'), tensor(1.8933e+08, device='cuda:0'), tensor(1.9217e+08, device='cuda:0'), tensor(1.8970e+08, device='cuda:0'), tensor(1.9339e+08, device='cuda:0')]

Training loss: 192202704.0

Prediction time: [datetime.timedelta(seconds=2, microseconds=39419), datetime.timedelta(seconds=2, microseconds=71290), datetime.timedelta(seconds=2, microseconds=58343), datetime.timedelta(seconds=2, microseconds=75269), datetime.timedelta(seconds=2, microseconds=59338), datetime.timedelta(seconds=2, microseconds=61331), datetime.timedelta(seconds=2, microseconds=54360), datetime.timedelta(seconds=2, microseconds=79255), datetime.timedelta(seconds=2, microseconds=63322), datetime.timedelta(seconds=2, microseconds=53364)]

Phi time: [datetime.timedelta(seconds=1, microseconds=595759), datetime.timedelta(microseconds=999067), datetime.timedelta(microseconds=959635), datetime.timedelta(microseconds=960401), datetime.timedelta(microseconds=961143), datetime.timedelta(microseconds=970823), datetime.timedelta(microseconds=964218), datetime.timedelta(microseconds=963391), datetime.timedelta(microseconds=964774), datetime.timedelta(microseconds=958388)]

