Precision: [tensor(0.0669, device='cuda:0'), tensor(0.2232, device='cuda:0'), tensor(0.2047, device='cuda:0'), tensor(0.2095, device='cuda:0'), tensor(0.1903, device='cuda:0'), tensor(0.1087, device='cuda:0'), tensor(0.1803, device='cuda:0'), tensor(0.2194, device='cuda:0'), tensor(0.2443, device='cuda:0'), tensor(0.1848, device='cuda:0')]

Output distance: [tensor(3.5543e+24, device='cuda:0'), tensor(7.4459e+22, device='cuda:0'), tensor(2.9162e+22, device='cuda:0'), tensor(1.2149e+22, device='cuda:0'), tensor(7.7661e+22, device='cuda:0'), tensor(1.3272e+23, device='cuda:0'), tensor(6.8400e+22, device='cuda:0'), tensor(1.3761e+23, device='cuda:0'), tensor(8.1344e+21, device='cuda:0'), tensor(1.2672e+22, device='cuda:0')]

Prediction loss: [tensor(5.4200e+24, device='cuda:0'), tensor(1.3907e+23, device='cuda:0'), tensor(5.8459e+22, device='cuda:0'), tensor(1.9703e+22, device='cuda:0'), tensor(1.4226e+23, device='cuda:0'), tensor(2.2437e+23, device='cuda:0'), tensor(1.0922e+23, device='cuda:0'), tensor(2.3254e+23, device='cuda:0'), tensor(1.3641e+22, device='cuda:0'), tensor(2.3141e+22, 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(17999, 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(9089392., device='cuda:0'), tensor(8282478., device='cuda:0'), tensor(8367097., device='cuda:0'), tensor(8940348., device='cuda:0'), tensor(8613275., device='cuda:0'), tensor(9644864., device='cuda:0'), tensor(8994705., device='cuda:0'), tensor(8486430., device='cuda:0'), tensor(8283543., device='cuda:0'), tensor(8808583., device='cuda:0')]

Training loss: 8893691.0

Prediction time: [datetime.timedelta(seconds=1, microseconds=364262), datetime.timedelta(seconds=1, microseconds=387116), datetime.timedelta(seconds=1, microseconds=366252), datetime.timedelta(seconds=1, microseconds=364260), datetime.timedelta(seconds=1, microseconds=361273), datetime.timedelta(seconds=1, microseconds=375213), datetime.timedelta(seconds=1, microseconds=413054), datetime.timedelta(seconds=1, microseconds=349326), datetime.timedelta(seconds=1, microseconds=394135), datetime.timedelta(seconds=1, microseconds=331398)]

Phi time: [datetime.timedelta(seconds=1, microseconds=272862), datetime.timedelta(microseconds=749356), datetime.timedelta(microseconds=665379), datetime.timedelta(microseconds=662479), datetime.timedelta(microseconds=685119), datetime.timedelta(microseconds=670148), datetime.timedelta(microseconds=670024), datetime.timedelta(microseconds=692037), datetime.timedelta(microseconds=669811), datetime.timedelta(microseconds=672630)]

