Precision: [tensor(0.2684, device='cuda:0'), tensor(0.2500, device='cuda:0'), tensor(0.2822, device='cuda:0'), tensor(0.2237, device='cuda:0'), tensor(0.2749, device='cuda:0'), tensor(0.2816, device='cuda:0'), tensor(0.2839, device='cuda:0'), tensor(0.2890, device='cuda:0'), tensor(0.2476, device='cuda:0'), tensor(0.2580, device='cuda:0')]

Output distance: [tensor(19.4885, device='cuda:0'), tensor(19.5254, device='cuda:0'), tensor(19.4610, device='cuda:0'), tensor(19.5780, device='cuda:0'), tensor(19.4755, device='cuda:0'), tensor(19.4622, device='cuda:0'), tensor(19.4577, device='cuda:0'), tensor(19.4474, device='cuda:0'), tensor(19.5302, device='cuda:0'), tensor(19.5094, device='cuda:0')]

Prediction loss: [tensor(107.9680, device='cuda:0'), tensor(108.2824, device='cuda:0'), tensor(107.4455, device='cuda:0'), tensor(108.0992, device='cuda:0'), tensor(109.2088, device='cuda:0'), tensor(106.8373, device='cuda:0'), tensor(107.3122, device='cuda:0'), tensor(108.8605, device='cuda:0'), tensor(107.6992, device='cuda:0'), tensor(107.1635, device='cuda:0')]

Others: [{'iter_num': 5, 'num_positive': tensor(6616, device='cuda:0'), 'num_positive_true': tensor(125872, device='cuda:0')}, {'iter_num': 5, 'num_positive': tensor(6616, device='cuda:0'), 'num_positive_true': tensor(125872, device='cuda:0')}, {'iter_num': 5, 'num_positive': tensor(6616, device='cuda:0'), 'num_positive_true': tensor(125872, device='cuda:0')}, {'iter_num': 5, 'num_positive': tensor(6616, device='cuda:0'), 'num_positive_true': tensor(125872, device='cuda:0')}, {'iter_num': 5, 'num_positive': tensor(6616, device='cuda:0'), 'num_positive_true': tensor(125872, device='cuda:0')}, {'iter_num': 5, 'num_positive': tensor(6616, device='cuda:0'), 'num_positive_true': tensor(125872, device='cuda:0')}, {'iter_num': 5, 'num_positive': tensor(6616, device='cuda:0'), 'num_positive_true': tensor(125872, device='cuda:0')}, {'iter_num': 5, 'num_positive': tensor(6616, device='cuda:0'), 'num_positive_true': tensor(125872, device='cuda:0')}, {'iter_num': 5, 'num_positive': tensor(6616, device='cuda:0'), 'num_positive_true': tensor(125872, device='cuda:0')}, {'iter_num': 5, 'num_positive': tensor(6616, 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=2, microseconds=557159), datetime.timedelta(seconds=2, microseconds=571097), datetime.timedelta(seconds=2, microseconds=577069), datetime.timedelta(seconds=2, microseconds=540228), datetime.timedelta(seconds=2, microseconds=319166), datetime.timedelta(seconds=2, microseconds=278339), datetime.timedelta(seconds=2, microseconds=256430), datetime.timedelta(seconds=2, microseconds=286303), datetime.timedelta(seconds=2, microseconds=260415), datetime.timedelta(seconds=2, microseconds=253443)]

Phi time: [datetime.timedelta(seconds=5, microseconds=466324), datetime.timedelta(seconds=4, microseconds=828239), datetime.timedelta(seconds=4, microseconds=486970), datetime.timedelta(seconds=4, microseconds=744462), datetime.timedelta(seconds=4, microseconds=750976), datetime.timedelta(seconds=4, microseconds=139215), datetime.timedelta(seconds=4, microseconds=89247), datetime.timedelta(seconds=4, microseconds=94401), datetime.timedelta(seconds=4, microseconds=111013), datetime.timedelta(seconds=4, microseconds=126946)]

