Precision: [tensor(0.8304, device='cuda:0'), tensor(0.8303, device='cuda:0'), tensor(0.8306, device='cuda:0'), tensor(0.8300, device='cuda:0'), tensor(0.8302, device='cuda:0'), tensor(0.8299, device='cuda:0'), tensor(0.8298, device='cuda:0'), tensor(0.8303, device='cuda:0'), tensor(0.8298, device='cuda:0'), tensor(0.8302, device='cuda:0')]

Output distance: [tensor(13433.7334, device='cuda:0'), tensor(13319.4355, device='cuda:0'), tensor(13262.3955, device='cuda:0'), tensor(13553.7256, device='cuda:0'), tensor(13355.1182, device='cuda:0'), tensor(13482.1416, device='cuda:0'), tensor(13715.9785, device='cuda:0'), tensor(13363.3496, device='cuda:0'), tensor(13438.1523, device='cuda:0'), tensor(13431.9531, device='cuda:0')]

Prediction loss: [tensor(10603.5400, device='cuda:0'), tensor(10506.7803, device='cuda:0'), tensor(10494.4443, device='cuda:0'), tensor(10801.6113, device='cuda:0'), tensor(10606.2236, device='cuda:0'), tensor(10775.9551, device='cuda:0'), tensor(10864.4775, device='cuda:0'), tensor(10566.1074, device='cuda:0'), tensor(10621.9893, device='cuda:0'), tensor(10717.2539, 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': 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': 11, '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(1.9338e+08, device='cuda:0'), tensor(1.9213e+08, device='cuda:0'), tensor(1.9206e+08, device='cuda:0'), tensor(1.9555e+08, device='cuda:0'), tensor(1.9354e+08, device='cuda:0'), tensor(1.9558e+08, device='cuda:0'), tensor(1.9588e+08, device='cuda:0'), tensor(1.9269e+08, device='cuda:0'), tensor(1.9352e+08, device='cuda:0'), tensor(1.9482e+08, device='cuda:0')]

Training loss: 193265312.0

Prediction time: [datetime.timedelta(seconds=1, microseconds=398120), datetime.timedelta(seconds=1, microseconds=253693), datetime.timedelta(seconds=2, microseconds=830991), datetime.timedelta(seconds=2, microseconds=829997), datetime.timedelta(seconds=2, microseconds=832986), datetime.timedelta(seconds=1, microseconds=251690), datetime.timedelta(seconds=2, microseconds=831991), datetime.timedelta(seconds=2, microseconds=825022), datetime.timedelta(seconds=2, microseconds=831988), datetime.timedelta(seconds=2, microseconds=834976)]

Phi time: [datetime.timedelta(seconds=1, microseconds=954214), datetime.timedelta(seconds=1, microseconds=287167), datetime.timedelta(seconds=1, microseconds=295486), datetime.timedelta(seconds=1, microseconds=294625), datetime.timedelta(seconds=1, microseconds=301256), datetime.timedelta(seconds=1, microseconds=322908), datetime.timedelta(seconds=1, microseconds=299081), datetime.timedelta(seconds=1, microseconds=293468), datetime.timedelta(seconds=1, microseconds=303727), datetime.timedelta(seconds=1, microseconds=299268)]

