Precision: [tensor(0.9998, device='cuda:0'), tensor(0.9998, device='cuda:0'), tensor(0.9998, device='cuda:0'), tensor(0.9998, device='cuda:0'), tensor(0.9998, device='cuda:0'), tensor(0.9998, device='cuda:0'), tensor(0.9998, device='cuda:0'), tensor(0.9998, device='cuda:0'), tensor(0.9998, device='cuda:0'), tensor(0.9998, device='cuda:0')]

Output distance: [tensor(140565., device='cuda:0'), tensor(138938.5000, device='cuda:0'), tensor(144448.6094, device='cuda:0'), tensor(138943.2656, device='cuda:0'), tensor(139138.7031, device='cuda:0'), tensor(139557.1250, device='cuda:0'), tensor(139514.0312, device='cuda:0'), tensor(141120.7344, device='cuda:0'), tensor(140988.5000, device='cuda:0'), tensor(141435.3438, device='cuda:0')]

Prediction loss: [tensor(138282.9219, device='cuda:0'), tensor(135446.7188, device='cuda:0'), tensor(143270.7656, device='cuda:0'), tensor(135852.6875, device='cuda:0'), tensor(136035.9375, device='cuda:0'), tensor(134929.4531, device='cuda:0'), tensor(135915.2031, device='cuda:0'), tensor(135268.3438, device='cuda:0'), tensor(140798.8281, device='cuda:0'), tensor(137693.6875, device='cuda:0')]

Others: [{'iter_num': 21, 'num_positive': tensor(6000, device='cuda:0'), 'num_positive_true': tensor(18000, device='cuda:0')}, {'iter_num': 11, 'num_positive': tensor(6000, device='cuda:0'), 'num_positive_true': tensor(18000, device='cuda:0')}, {'iter_num': 29, 'num_positive': tensor(6000, device='cuda:0'), 'num_positive_true': tensor(18000, device='cuda:0')}, {'iter_num': 13, 'num_positive': tensor(6000, device='cuda:0'), 'num_positive_true': tensor(18000, device='cuda:0')}, {'iter_num': 7, 'num_positive': tensor(6000, device='cuda:0'), 'num_positive_true': tensor(18000, device='cuda:0')}, {'iter_num': 30, 'num_positive': tensor(6000, device='cuda:0'), 'num_positive_true': tensor(18000, device='cuda:0')}, {'iter_num': 17, 'num_positive': tensor(6000, device='cuda:0'), 'num_positive_true': tensor(18000, device='cuda:0')}, {'iter_num': 23, 'num_positive': tensor(6000, device='cuda:0'), 'num_positive_true': tensor(18000, device='cuda:0')}, {'iter_num': 23, 'num_positive': tensor(6000, device='cuda:0'), 'num_positive_true': tensor(18000, device='cuda:0')}, {'iter_num': 23, 'num_positive': tensor(6000, device='cuda:0'), 'num_positive_true': tensor(18000, device='cuda:0')}]

Compressed training loss: [tensor(1.9174e+08, device='cuda:0'), tensor(1.9124e+08, device='cuda:0'), tensor(1.9309e+08, device='cuda:0'), tensor(1.9100e+08, device='cuda:0'), tensor(1.9067e+08, device='cuda:0'), tensor(1.8957e+08, device='cuda:0'), tensor(1.9095e+08, device='cuda:0'), tensor(1.9019e+08, device='cuda:0'), tensor(1.9323e+08, device='cuda:0'), tensor(1.9168e+08, device='cuda:0')]

Training loss: 191944544.0

Prediction time: [datetime.timedelta(seconds=1, microseconds=933859), datetime.timedelta(seconds=1, microseconds=189989), datetime.timedelta(seconds=2, microseconds=574164), datetime.timedelta(seconds=1, microseconds=331392), datetime.timedelta(microseconds=862370), datetime.timedelta(seconds=2, microseconds=648849), datetime.timedelta(seconds=1, microseconds=653043), datetime.timedelta(seconds=2, microseconds=100160), datetime.timedelta(seconds=2, microseconds=124058), datetime.timedelta(seconds=2, microseconds=118084)]

Phi time: [datetime.timedelta(seconds=1, microseconds=873911), datetime.timedelta(seconds=1, microseconds=263177), datetime.timedelta(seconds=1, microseconds=270599), datetime.timedelta(seconds=1, microseconds=275158), datetime.timedelta(seconds=1, microseconds=266197), datetime.timedelta(seconds=1, microseconds=272389), datetime.timedelta(seconds=1, microseconds=277120), datetime.timedelta(seconds=1, microseconds=274577), datetime.timedelta(seconds=1, microseconds=277623), datetime.timedelta(seconds=1, microseconds=297537)]

