Precision: [tensor(0.0305, device='cuda:0'), tensor(0.0287, device='cuda:0'), tensor(0.0352, device='cuda:0'), tensor(0.0328, device='cuda:0'), tensor(0.0318, device='cuda:0'), tensor(0.0311, device='cuda:0'), tensor(0.0301, device='cuda:0'), tensor(0.0330, device='cuda:0'), tensor(0.0311, device='cuda:0'), tensor(0.0363, device='cuda:0')]

Output distance: [tensor(21.8422, device='cuda:0'), tensor(21.8534, device='cuda:0'), tensor(21.8144, device='cuda:0'), tensor(21.8286, device='cuda:0'), tensor(21.8343, device='cuda:0'), tensor(21.8386, device='cuda:0'), tensor(21.8449, device='cuda:0'), tensor(21.8277, device='cuda:0'), tensor(21.8389, device='cuda:0'), tensor(21.8077, device='cuda:0')]

Prediction loss: [tensor(112.6315, device='cuda:0'), tensor(114.8514, device='cuda:0'), tensor(113.7097, device='cuda:0'), tensor(113.0683, device='cuda:0'), tensor(113.3850, device='cuda:0'), tensor(113.2245, device='cuda:0'), tensor(114.4089, device='cuda:0'), tensor(114.7563, device='cuda:0'), tensor(112.8425, device='cuda:0'), tensor(111.9116, 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=650618), datetime.timedelta(seconds=3, microseconds=647631), datetime.timedelta(seconds=3, microseconds=668548), datetime.timedelta(seconds=3, microseconds=656597), datetime.timedelta(seconds=3, microseconds=660576), datetime.timedelta(seconds=3, microseconds=664560), datetime.timedelta(seconds=3, microseconds=643648), datetime.timedelta(seconds=3, microseconds=653605), datetime.timedelta(seconds=3, microseconds=651614), datetime.timedelta(seconds=3, microseconds=669437)]

Phi time: [datetime.timedelta(seconds=4, microseconds=288060), datetime.timedelta(seconds=4, microseconds=128634), datetime.timedelta(seconds=4, microseconds=139862), datetime.timedelta(seconds=4, microseconds=99093), datetime.timedelta(seconds=4, microseconds=87181), datetime.timedelta(seconds=4, microseconds=91438), datetime.timedelta(seconds=4, microseconds=93118), datetime.timedelta(seconds=4, microseconds=84788), datetime.timedelta(seconds=4, microseconds=94901), datetime.timedelta(seconds=4, microseconds=97150)]

