Precision: [tensor(0.9588, device='cuda:0'), tensor(0.9533, device='cuda:0'), tensor(0.9647, device='cuda:0'), tensor(0.9571, device='cuda:0'), tensor(0.9632, device='cuda:0'), tensor(0.9532, device='cuda:0'), tensor(0.9581, device='cuda:0'), tensor(0.9567, device='cuda:0'), tensor(0.9545, device='cuda:0'), tensor(0.9587, device='cuda:0')]

Output distance: [tensor(112.0858, device='cuda:0'), tensor(115.5131, device='cuda:0'), tensor(85.9782, device='cuda:0'), tensor(110.1049, device='cuda:0'), tensor(92.7220, device='cuda:0'), tensor(122.1620, device='cuda:0'), tensor(107.0370, device='cuda:0'), tensor(110.7904, device='cuda:0'), tensor(118.1596, device='cuda:0'), tensor(104.4183, device='cuda:0')]

Prediction loss: [tensor(391.1934, device='cuda:0'), tensor(398.7035, device='cuda:0'), tensor(340.4544, device='cuda:0'), tensor(362.2651, device='cuda:0'), tensor(367.3076, device='cuda:0'), tensor(369.5750, device='cuda:0'), tensor(370.8277, device='cuda:0'), tensor(377.4293, device='cuda:0'), tensor(381.9921, device='cuda:0'), tensor(356.3528, device='cuda:0')]

Others: [{'iter_num': 15, 'num_positive': tensor(18000, device='cuda:0'), 'num_positive_true': tensor(18000, device='cuda:0')}, {'iter_num': 13, 'num_positive': tensor(18000, device='cuda:0'), 'num_positive_true': tensor(18000, device='cuda:0')}, {'iter_num': 15, 'num_positive': tensor(18000, device='cuda:0'), 'num_positive_true': tensor(18000, device='cuda:0')}, {'iter_num': 13, 'num_positive': tensor(18000, device='cuda:0'), 'num_positive_true': tensor(18000, device='cuda:0')}, {'iter_num': 15, 'num_positive': tensor(18000, device='cuda:0'), 'num_positive_true': tensor(18000, device='cuda:0')}, {'iter_num': 13, 'num_positive': tensor(18000, device='cuda:0'), 'num_positive_true': tensor(18000, device='cuda:0')}, {'iter_num': 15, 'num_positive': tensor(18000, device='cuda:0'), 'num_positive_true': tensor(18000, device='cuda:0')}, {'iter_num': 13, 'num_positive': tensor(18000, device='cuda:0'), 'num_positive_true': tensor(18000, device='cuda:0')}, {'iter_num': 15, 'num_positive': tensor(18000, device='cuda:0'), 'num_positive_true': tensor(18000, device='cuda:0')}, {'iter_num': 17, 'num_positive': tensor(18000, device='cuda:0'), 'num_positive_true': tensor(18000, device='cuda:0')}]

Compressed training loss: [tensor(3768254.5000, device='cuda:0'), tensor(3833147.2500, device='cuda:0'), tensor(3222292.2500, device='cuda:0'), tensor(3471462.7500, device='cuda:0'), tensor(3502003.2500, device='cuda:0'), tensor(3567527.2500, device='cuda:0'), tensor(3569750.2500, device='cuda:0'), tensor(3631326.2500, device='cuda:0'), tensor(3716943.5000, device='cuda:0'), tensor(3423538., device='cuda:0')]

Training loss: 3596177.0

Prediction time: [datetime.timedelta(microseconds=763761), datetime.timedelta(microseconds=736878), datetime.timedelta(microseconds=789650), datetime.timedelta(microseconds=717955), datetime.timedelta(microseconds=787660), datetime.timedelta(microseconds=716959), datetime.timedelta(microseconds=800607), datetime.timedelta(microseconds=717954), datetime.timedelta(microseconds=789651), datetime.timedelta(microseconds=842426)]

Phi time: [datetime.timedelta(seconds=1, microseconds=319557), datetime.timedelta(microseconds=811841), datetime.timedelta(microseconds=738933), datetime.timedelta(microseconds=743055), datetime.timedelta(microseconds=751013), datetime.timedelta(microseconds=739163), datetime.timedelta(microseconds=769486), datetime.timedelta(microseconds=741069), datetime.timedelta(microseconds=738200), datetime.timedelta(microseconds=742502)]

