Precision: [tensor(0.3463, device='cuda:0'), tensor(0.3422, device='cuda:0'), tensor(0.3423, device='cuda:0'), tensor(0.3435, device='cuda:0'), tensor(0.3450, device='cuda:0'), tensor(0.3453, device='cuda:0'), tensor(0.3473, device='cuda:0'), tensor(0.3460, device='cuda:0'), tensor(0.3431, device='cuda:0'), tensor(0.3479, device='cuda:0')]

Output distance: [tensor(20.5626, device='cuda:0'), tensor(20.6031, device='cuda:0'), tensor(20.6025, device='cuda:0'), tensor(20.5904, device='cuda:0'), tensor(20.5753, device='cuda:0'), tensor(20.5729, device='cuda:0'), tensor(20.5523, device='cuda:0'), tensor(20.5650, device='cuda:0'), tensor(20.5940, device='cuda:0'), tensor(20.5466, device='cuda:0')]

Prediction loss: [tensor(102.6439, device='cuda:0'), tensor(102.5480, device='cuda:0'), tensor(103.4443, device='cuda:0'), tensor(102.1482, device='cuda:0'), tensor(101.4313, device='cuda:0'), tensor(102.0652, device='cuda:0'), tensor(102.2935, device='cuda:0'), tensor(103.0442, device='cuda:0'), tensor(102.1400, device='cuda:0'), tensor(102.0345, device='cuda:0')]

Others: [{'iter_num': 60, 'num_positive': tensor(33080, device='cuda:0'), 'num_positive_true': tensor(125872, device='cuda:0')}, {'iter_num': 60, 'num_positive': tensor(33080, device='cuda:0'), 'num_positive_true': tensor(125872, device='cuda:0')}, {'iter_num': 60, 'num_positive': tensor(33080, device='cuda:0'), 'num_positive_true': tensor(125872, device='cuda:0')}, {'iter_num': 60, 'num_positive': tensor(33080, device='cuda:0'), 'num_positive_true': tensor(125872, device='cuda:0')}, {'iter_num': 60, 'num_positive': tensor(33080, device='cuda:0'), 'num_positive_true': tensor(125872, device='cuda:0')}, {'iter_num': 60, 'num_positive': tensor(33080, device='cuda:0'), 'num_positive_true': tensor(125872, device='cuda:0')}, {'iter_num': 60, 'num_positive': tensor(33080, device='cuda:0'), 'num_positive_true': tensor(125872, device='cuda:0')}, {'iter_num': 60, 'num_positive': tensor(33080, device='cuda:0'), 'num_positive_true': tensor(125872, device='cuda:0')}, {'iter_num': 60, 'num_positive': tensor(33080, device='cuda:0'), 'num_positive_true': tensor(125872, device='cuda:0')}, {'iter_num': 60, 'num_positive': tensor(33080, device='cuda:0'), 'num_positive_true': tensor(125872, device='cuda:0')}]

Compressed training loss: [0, 0, 0, 0, 0, 0, 0, 0, 0, 0]

Training loss: 0

Prediction time: [datetime.timedelta(seconds=8, microseconds=301980), datetime.timedelta(seconds=8, microseconds=298872), datetime.timedelta(seconds=8, microseconds=277892), datetime.timedelta(seconds=8, microseconds=297809), datetime.timedelta(seconds=8, microseconds=308760), datetime.timedelta(seconds=8, microseconds=331667), datetime.timedelta(seconds=8, microseconds=311798), datetime.timedelta(seconds=8, microseconds=310750), datetime.timedelta(seconds=8, microseconds=305771), datetime.timedelta(seconds=8, microseconds=316722)]

Phi time: [datetime.timedelta(seconds=5, microseconds=596285), datetime.timedelta(seconds=5, microseconds=621464), datetime.timedelta(seconds=5, microseconds=637992), datetime.timedelta(seconds=5, microseconds=651600), datetime.timedelta(seconds=5, microseconds=641660), datetime.timedelta(seconds=5, microseconds=626633), datetime.timedelta(seconds=5, microseconds=644186), datetime.timedelta(seconds=5, microseconds=647467), datetime.timedelta(seconds=5, microseconds=677710), datetime.timedelta(seconds=5, microseconds=662618)]

