Precision: [tensor(0.0223, device='cuda:0'), tensor(0.0280, device='cuda:0'), tensor(0.0343, device='cuda:0'), tensor(0.0284, device='cuda:0'), tensor(0.0302, device='cuda:0'), tensor(0.0404, device='cuda:0'), tensor(0.0278, device='cuda:0'), tensor(0.0325, device='cuda:0'), tensor(0.0359, device='cuda:0'), tensor(0.0204, device='cuda:0')]

Output distance: [tensor(21.8915, device='cuda:0'), tensor(21.8573, device='cuda:0'), tensor(21.8195, device='cuda:0'), tensor(21.8552, device='cuda:0'), tensor(21.8440, device='cuda:0'), tensor(21.7833, device='cuda:0'), tensor(21.8588, device='cuda:0'), tensor(21.8304, device='cuda:0'), tensor(21.8102, device='cuda:0'), tensor(21.9030, device='cuda:0')]

Prediction loss: [tensor(102.3789, device='cuda:0'), tensor(105.0134, device='cuda:0'), tensor(105.1149, device='cuda:0'), tensor(101.2169, device='cuda:0'), tensor(100.5618, device='cuda:0'), tensor(104.8595, device='cuda:0'), tensor(100.7371, device='cuda:0'), tensor(103.0445, device='cuda:0'), tensor(103.2596, device='cuda:0'), tensor(103.0936, device='cuda:0')]

Others: [{'iter_num': 30, 'num_positive': tensor(19848, device='cuda:0'), 'num_positive_true': tensor(125872, device='cuda:0')}, {'iter_num': 30, 'num_positive': tensor(19848, device='cuda:0'), 'num_positive_true': tensor(125872, device='cuda:0')}, {'iter_num': 30, 'num_positive': tensor(19848, device='cuda:0'), 'num_positive_true': tensor(125872, device='cuda:0')}, {'iter_num': 30, 'num_positive': tensor(19848, device='cuda:0'), 'num_positive_true': tensor(125872, device='cuda:0')}, {'iter_num': 30, 'num_positive': tensor(19848, device='cuda:0'), 'num_positive_true': tensor(125872, device='cuda:0')}, {'iter_num': 30, 'num_positive': tensor(19848, device='cuda:0'), 'num_positive_true': tensor(125872, device='cuda:0')}, {'iter_num': 30, 'num_positive': tensor(19848, device='cuda:0'), 'num_positive_true': tensor(125872, device='cuda:0')}, {'iter_num': 30, 'num_positive': tensor(19848, device='cuda:0'), 'num_positive_true': tensor(125872, device='cuda:0')}, {'iter_num': 30, 'num_positive': tensor(19848, device='cuda:0'), 'num_positive_true': tensor(125872, device='cuda:0')}, {'iter_num': 30, 'num_positive': tensor(19848, device='cuda:0'), 'num_positive_true': tensor(125872, device='cuda:0')}]

Compressed training loss: [tensor(0.0001, device='cuda:0'), tensor(0.0001, device='cuda:0'), tensor(0.0001, device='cuda:0'), tensor(0.0001, device='cuda:0'), tensor(0.0001, device='cuda:0'), tensor(0.0001, device='cuda:0'), tensor(0.0001, device='cuda:0'), tensor(0.0001, device='cuda:0'), tensor(0.0001, device='cuda:0'), tensor(0.0001, device='cuda:0')]

Training loss: 0

Prediction time: [datetime.timedelta(seconds=3, microseconds=122574), datetime.timedelta(seconds=3, microseconds=137973), datetime.timedelta(seconds=3, microseconds=128467), datetime.timedelta(seconds=3, microseconds=131687), datetime.timedelta(seconds=3, microseconds=132429), datetime.timedelta(seconds=3, microseconds=126549), datetime.timedelta(seconds=3, microseconds=123245), datetime.timedelta(seconds=3, microseconds=126306), datetime.timedelta(seconds=3, microseconds=180424), datetime.timedelta(seconds=3, microseconds=121716)]

Phi time: [datetime.timedelta(seconds=170, microseconds=598981), datetime.timedelta(seconds=170, microseconds=493136), datetime.timedelta(seconds=170, microseconds=652010), datetime.timedelta(seconds=170, microseconds=635181), datetime.timedelta(seconds=170, microseconds=293783), datetime.timedelta(seconds=170, microseconds=370010), datetime.timedelta(seconds=170, microseconds=300209), datetime.timedelta(seconds=170, microseconds=401338), datetime.timedelta(seconds=170, microseconds=443344), datetime.timedelta(seconds=170, microseconds=490720)]

