Precision: [tensor(0.8146, device='cuda:0'), tensor(0.8086, device='cuda:0'), tensor(0.8086, device='cuda:0'), tensor(0.8106, device='cuda:0'), tensor(0.8102, device='cuda:0'), tensor(0.8144, device='cuda:0'), tensor(0.8124, device='cuda:0'), tensor(0.8157, device='cuda:0'), tensor(0.8116, device='cuda:0'), tensor(0.8104, device='cuda:0')]

Output distance: [tensor(18633.4668, device='cuda:0'), tensor(18269.5352, device='cuda:0'), tensor(18159.6230, device='cuda:0'), tensor(17546.4629, device='cuda:0'), tensor(18324.5000, device='cuda:0'), tensor(18190.6777, device='cuda:0'), tensor(20507.1543, device='cuda:0'), tensor(17333.7812, device='cuda:0'), tensor(19210.8320, device='cuda:0'), tensor(18836.3984, device='cuda:0')]

Prediction loss: [tensor(16609.5703, device='cuda:0'), tensor(16124.2041, device='cuda:0'), tensor(16106.0127, device='cuda:0'), tensor(15847.6729, device='cuda:0'), tensor(15955.6748, device='cuda:0'), tensor(16157.7510, device='cuda:0'), tensor(17427.1484, device='cuda:0'), tensor(14762.8691, device='cuda:0'), tensor(16835.8789, device='cuda:0'), tensor(17082.9180, device='cuda:0')]

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

Compressed training loss: [tensor(1.9042e+08, device='cuda:0'), tensor(1.9063e+08, device='cuda:0'), tensor(1.9267e+08, device='cuda:0'), tensor(1.9282e+08, device='cuda:0'), tensor(1.8970e+08, device='cuda:0'), tensor(1.9262e+08, device='cuda:0'), tensor(1.9221e+08, device='cuda:0'), tensor(1.9027e+08, device='cuda:0'), tensor(1.9257e+08, device='cuda:0'), tensor(1.9277e+08, device='cuda:0')]

Training loss: 191466656.0

Prediction time: [datetime.timedelta(microseconds=128464), datetime.timedelta(microseconds=127463), datetime.timedelta(microseconds=123482), datetime.timedelta(microseconds=123480), datetime.timedelta(microseconds=126519), datetime.timedelta(microseconds=125472), datetime.timedelta(microseconds=129458), datetime.timedelta(microseconds=128460), datetime.timedelta(microseconds=125472), datetime.timedelta(microseconds=126468)]

Phi time: [datetime.timedelta(seconds=1, microseconds=960231), datetime.timedelta(seconds=1, microseconds=310892), datetime.timedelta(seconds=1, microseconds=266756), datetime.timedelta(seconds=1, microseconds=324502), datetime.timedelta(seconds=1, microseconds=351520), datetime.timedelta(seconds=1, microseconds=325095), datetime.timedelta(seconds=1, microseconds=315829), datetime.timedelta(seconds=1, microseconds=340036), datetime.timedelta(seconds=1, microseconds=308545), datetime.timedelta(seconds=1, microseconds=315937)]

