Precision: [tensor(0.8558, device='cuda:0'), tensor(0.8557, device='cuda:0'), tensor(0.8567, device='cuda:0'), tensor(0.8567, device='cuda:0'), tensor(0.8557, device='cuda:0'), tensor(0.8543, device='cuda:0'), tensor(0.8546, device='cuda:0'), tensor(0.8563, device='cuda:0'), tensor(0.8557, device='cuda:0'), tensor(0.8546, device='cuda:0')]

Output distance: [tensor(553.1801, device='cuda:0'), tensor(586.3184, device='cuda:0'), tensor(562.4559, device='cuda:0'), tensor(549.9108, device='cuda:0'), tensor(554.9786, device='cuda:0'), tensor(563.3439, device='cuda:0'), tensor(569.6249, device='cuda:0'), tensor(554.0714, device='cuda:0'), tensor(574.5641, device='cuda:0'), tensor(562.5722, device='cuda:0')]

Prediction loss: [tensor(598.9573, device='cuda:0'), tensor(641.9559, device='cuda:0'), tensor(632.7692, device='cuda:0'), tensor(593.5882, device='cuda:0'), tensor(614.4215, device='cuda:0'), tensor(600.9854, device='cuda:0'), tensor(613.0351, device='cuda:0'), tensor(612.1179, device='cuda:0'), tensor(643.6846, device='cuda:0'), tensor(609.5593, 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(8655535., device='cuda:0'), tensor(8984935., device='cuda:0'), tensor(8950115., device='cuda:0'), tensor(8603576., device='cuda:0'), tensor(8835062., device='cuda:0'), tensor(8657585., device='cuda:0'), tensor(8801293., device='cuda:0'), tensor(8792923., device='cuda:0'), tensor(9070945., device='cuda:0'), tensor(8780842., device='cuda:0')]

Training loss: 8837783.0

Prediction time: [datetime.timedelta(seconds=1, microseconds=817292), datetime.timedelta(seconds=1, microseconds=835216), datetime.timedelta(seconds=1, microseconds=814306), datetime.timedelta(seconds=1, microseconds=836212), datetime.timedelta(seconds=1, microseconds=849157), datetime.timedelta(seconds=1, microseconds=827254), datetime.timedelta(seconds=1, microseconds=816298), datetime.timedelta(seconds=1, microseconds=832174), datetime.timedelta(seconds=1, microseconds=814306), datetime.timedelta(seconds=1, microseconds=815301)]

Phi time: [datetime.timedelta(seconds=1, microseconds=502064), datetime.timedelta(microseconds=932229), datetime.timedelta(microseconds=868143), datetime.timedelta(microseconds=868324), datetime.timedelta(microseconds=868378), datetime.timedelta(microseconds=873000), datetime.timedelta(microseconds=868702), datetime.timedelta(microseconds=867526), datetime.timedelta(microseconds=869979), datetime.timedelta(microseconds=869171)]

