Precision: [tensor(0.8451, device='cuda:0'), tensor(0.8391, device='cuda:0'), tensor(0.8509, device='cuda:0'), tensor(0.8504, device='cuda:0'), tensor(0.8481, device='cuda:0'), tensor(0.8515, device='cuda:0'), tensor(0.8331, device='cuda:0'), tensor(0.8329, device='cuda:0'), tensor(0.8345, device='cuda:0'), tensor(0.8517, device='cuda:0')]

Output distance: [tensor(594.9750, device='cuda:0'), tensor(682.8928, device='cuda:0'), tensor(599.5033, device='cuda:0'), tensor(560.2731, device='cuda:0'), tensor(636.9424, device='cuda:0'), tensor(560.2167, device='cuda:0'), tensor(912.8932, device='cuda:0'), tensor(1992.4447, device='cuda:0'), tensor(905.8249, device='cuda:0'), tensor(581.7194, device='cuda:0')]

Prediction loss: [tensor(617.2180, device='cuda:0'), tensor(741.3193, device='cuda:0'), tensor(650.3210, device='cuda:0'), tensor(591.7755, device='cuda:0'), tensor(663.4656, device='cuda:0'), tensor(602.8986, device='cuda:0'), tensor(1017.8022, device='cuda:0'), tensor(2955.6799, device='cuda:0'), tensor(976.8485, device='cuda:0'), tensor(639.2675, 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(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')}, {'iter_num': 30, 'num_positive': tensor(18000, device='cuda:0'), 'num_positive_true': tensor(18000, device='cuda:0')}]

Compressed training loss: [tensor(8940418., device='cuda:0'), tensor(9116465., device='cuda:0'), tensor(9185364., device='cuda:0'), tensor(8883094., device='cuda:0'), tensor(8802431., device='cuda:0'), tensor(8927118., device='cuda:0'), tensor(9472504., device='cuda:0'), tensor(9853536., device='cuda:0'), tensor(8891035., device='cuda:0'), tensor(9030955., device='cuda:0')]

Training loss: 8925608.0

Prediction time: [datetime.timedelta(seconds=1, microseconds=135225), datetime.timedelta(seconds=1, microseconds=153149), datetime.timedelta(seconds=1, microseconds=132234), datetime.timedelta(seconds=1, microseconds=145229), datetime.timedelta(seconds=1, microseconds=131243), datetime.timedelta(seconds=1, microseconds=156136), datetime.timedelta(seconds=1, microseconds=151157), datetime.timedelta(seconds=1, microseconds=130245), datetime.timedelta(seconds=1, microseconds=135224), datetime.timedelta(seconds=1, microseconds=157132)]

Phi time: [datetime.timedelta(seconds=1, microseconds=217585), datetime.timedelta(microseconds=702345), datetime.timedelta(microseconds=624913), datetime.timedelta(microseconds=623393), datetime.timedelta(microseconds=631109), datetime.timedelta(microseconds=628443), datetime.timedelta(microseconds=637242), datetime.timedelta(microseconds=631609), datetime.timedelta(microseconds=624260), datetime.timedelta(microseconds=624727)]

