Special test with strict convergence condition

Precision: [tensor(0.6855, device='cuda:0'), tensor(0.6934, device='cuda:0'), tensor(0.6873, device='cuda:0'), tensor(0.6905, device='cuda:0'), tensor(0.6892, device='cuda:0'), tensor(0.6836, device='cuda:0'), tensor(0.6892, device='cuda:0'), tensor(0.6902, device='cuda:0'), tensor(0.6913, device='cuda:0'), tensor(0.6939, device='cuda:0')]

Output distance: [tensor(4.9352, device='cuda:0'), tensor(4.9194, device='cuda:0'), tensor(4.9315, device='cuda:0'), tensor(4.9252, device='cuda:0'), tensor(4.9278, device='cuda:0'), tensor(4.9388, device='cuda:0'), tensor(4.9278, device='cuda:0'), tensor(4.9257, device='cuda:0'), tensor(4.9236, device='cuda:0'), tensor(4.9184, device='cuda:0')]

Prediction loss: [tensor(17717336., device='cuda:0'), tensor(18537372., device='cuda:0'), tensor(17757432., device='cuda:0'), tensor(17971632., device='cuda:0'), tensor(17942776., device='cuda:0'), tensor(19604912., device='cuda:0'), tensor(17010986., device='cuda:0'), tensor(17927330., device='cuda:0'), tensor(18824456., device='cuda:0'), tensor(18256492., device='cuda:0')]

Others: [{'iter_num': 9, 'num_positive': tensor(3809, device='cuda:0'), 'num_positive_true': tensor(20211, device='cuda:0')}, {'iter_num': 7, 'num_positive': tensor(3809, device='cuda:0'), 'num_positive_true': tensor(20211, device='cuda:0')}, {'iter_num': 7, 'num_positive': tensor(3809, device='cuda:0'), 'num_positive_true': tensor(20211, device='cuda:0')}, {'iter_num': 9, 'num_positive': tensor(3809, device='cuda:0'), 'num_positive_true': tensor(20211, device='cuda:0')}, {'iter_num': 9, 'num_positive': tensor(3809, device='cuda:0'), 'num_positive_true': tensor(20211, device='cuda:0')}, {'iter_num': 9, 'num_positive': tensor(3809, device='cuda:0'), 'num_positive_true': tensor(20211, device='cuda:0')}, {'iter_num': 9, 'num_positive': tensor(3809, device='cuda:0'), 'num_positive_true': tensor(20211, device='cuda:0')}, {'iter_num': 9, 'num_positive': tensor(3809, device='cuda:0'), 'num_positive_true': tensor(20211, device='cuda:0')}, {'iter_num': 9, 'num_positive': tensor(3809, device='cuda:0'), 'num_positive_true': tensor(20211, device='cuda:0')}, {'iter_num': 9, 'num_positive': tensor(3809, device='cuda:0'), 'num_positive_true': tensor(20211, device='cuda:0')}]

Compressed training loss: [tensor(40940.3516, device='cuda:0'), tensor(40832.2266, device='cuda:0'), tensor(41024.4766, device='cuda:0'), tensor(40785.6445, device='cuda:0'), tensor(40807.4258, device='cuda:0'), tensor(40882.2578, device='cuda:0'), tensor(40935.1406, device='cuda:0'), tensor(40688.1133, device='cuda:0'), tensor(40732.5117, device='cuda:0'), tensor(40851.4883, device='cuda:0')]

Training loss: 0

Prediction time: [datetime.timedelta(seconds=1, microseconds=110291), datetime.timedelta(seconds=1, microseconds=50545), datetime.timedelta(seconds=1, microseconds=19674), datetime.timedelta(seconds=1, microseconds=36603), datetime.timedelta(seconds=1, microseconds=57515), datetime.timedelta(seconds=1, microseconds=52536), datetime.timedelta(seconds=1, microseconds=26645), datetime.timedelta(seconds=1, microseconds=22663), datetime.timedelta(seconds=1, microseconds=42579), datetime.timedelta(seconds=1, microseconds=48553)]

Phi time: [datetime.timedelta(microseconds=238986), datetime.timedelta(microseconds=250936), datetime.timedelta(microseconds=245958), datetime.timedelta(microseconds=250937), datetime.timedelta(microseconds=235999), datetime.timedelta(microseconds=234008), datetime.timedelta(microseconds=258903), datetime.timedelta(microseconds=230024), datetime.timedelta(microseconds=235003), datetime.timedelta(microseconds=232016)]

