Special test with strict convergence condition

Precision: [tensor(0.7292, device='cuda:0'), tensor(0.7318, device='cuda:0'), tensor(0.7296, device='cuda:0'), tensor(0.7364, device='cuda:0'), tensor(0.7306, device='cuda:0'), tensor(0.7339, device='cuda:0'), tensor(0.7271, device='cuda:0'), tensor(0.7352, device='cuda:0'), tensor(0.7363, device='cuda:0'), tensor(0.7353, device='cuda:0')]

Output distance: [tensor(5.0444, device='cuda:0'), tensor(5.0420, device='cuda:0'), tensor(5.0417, device='cuda:0'), tensor(5.0331, device='cuda:0'), tensor(5.0441, device='cuda:0'), tensor(5.0383, device='cuda:0'), tensor(5.0431, device='cuda:0'), tensor(5.0357, device='cuda:0'), tensor(5.0360, device='cuda:0'), tensor(5.0320, device='cuda:0')]

Prediction loss: [tensor(18805566., device='cuda:0'), tensor(17821840., device='cuda:0'), tensor(18873416., device='cuda:0'), tensor(17018964., device='cuda:0'), tensor(18805418., device='cuda:0'), tensor(19498296., device='cuda:0'), tensor(19395774., device='cuda:0'), tensor(20252926., device='cuda:0'), tensor(19006200., device='cuda:0'), tensor(18095418., device='cuda:0')]

Others: [{'iter_num': 5, 'num_positive': tensor(2175, device='cuda:0'), 'num_positive_true': tensor(20211, device='cuda:0')}, {'iter_num': 5, 'num_positive': tensor(2170, device='cuda:0'), 'num_positive_true': tensor(20211, device='cuda:0')}, {'iter_num': 5, 'num_positive': tensor(2193, device='cuda:0'), 'num_positive_true': tensor(20211, device='cuda:0')}, {'iter_num': 5, 'num_positive': tensor(2200, device='cuda:0'), 'num_positive_true': tensor(20211, device='cuda:0')}, {'iter_num': 5, 'num_positive': tensor(2164, device='cuda:0'), 'num_positive_true': tensor(20211, device='cuda:0')}, {'iter_num': 5, 'num_positive': tensor(2180, device='cuda:0'), 'num_positive_true': tensor(20211, device='cuda:0')}, {'iter_num': 5, 'num_positive': tensor(2206, device='cuda:0'), 'num_positive_true': tensor(20211, device='cuda:0')}, {'iter_num': 5, 'num_positive': tensor(2190, device='cuda:0'), 'num_positive_true': tensor(20211, device='cuda:0')}, {'iter_num': 5, 'num_positive': tensor(2177, device='cuda:0'), 'num_positive_true': tensor(20211, device='cuda:0')}, {'iter_num': 5, 'num_positive': tensor(2218, device='cuda:0'), 'num_positive_true': tensor(20211, device='cuda:0')}]

Compressed training loss: [tensor(40937.0469, device='cuda:0'), tensor(40807.0039, device='cuda:0'), tensor(40980.1328, device='cuda:0'), tensor(40887.4688, device='cuda:0'), tensor(40890.6133, device='cuda:0'), tensor(40801.3633, device='cuda:0'), tensor(40809.3711, device='cuda:0'), tensor(40929.8906, device='cuda:0'), tensor(40914.5000, device='cuda:0'), tensor(40921.5117, device='cuda:0')]

Training loss: 0

Prediction time: [datetime.timedelta(seconds=6, microseconds=228850), datetime.timedelta(seconds=6, microseconds=68523), datetime.timedelta(seconds=5, microseconds=942054), datetime.timedelta(seconds=6, microseconds=162130), datetime.timedelta(seconds=6, microseconds=91426), datetime.timedelta(seconds=6, microseconds=164122), datetime.timedelta(seconds=6, microseconds=145203), datetime.timedelta(seconds=6, microseconds=76491), datetime.timedelta(seconds=6, microseconds=168105), datetime.timedelta(seconds=6, microseconds=147193)]

Phi time: [datetime.timedelta(microseconds=357500), datetime.timedelta(microseconds=409282), datetime.timedelta(microseconds=447122), datetime.timedelta(microseconds=416252), datetime.timedelta(microseconds=387375), datetime.timedelta(microseconds=424219), datetime.timedelta(microseconds=385382), datetime.timedelta(microseconds=380403), datetime.timedelta(microseconds=357499), datetime.timedelta(microseconds=418244)]

