Special test with strict convergence condition

Precision: [tensor(0.7389, device='cuda:0'), tensor(0.7306, device='cuda:0'), tensor(0.7404, device='cuda:0'), tensor(0.7362, device='cuda:0'), tensor(0.7367, device='cuda:0'), tensor(0.7359, device='cuda:0'), tensor(0.7343, device='cuda:0'), tensor(0.7305, device='cuda:0'), tensor(0.7349, device='cuda:0'), tensor(0.7348, device='cuda:0')]

Output distance: [tensor(5.0063, device='cuda:0'), tensor(5.0163, device='cuda:0'), tensor(5.0047, device='cuda:0'), tensor(5.0089, device='cuda:0'), tensor(5.0079, device='cuda:0'), tensor(5.0084, device='cuda:0'), tensor(5.0108, device='cuda:0'), tensor(5.0147, device='cuda:0'), tensor(5.0097, device='cuda:0'), tensor(5.0123, device='cuda:0')]

Prediction loss: [tensor(17756464., device='cuda:0'), tensor(19214902., device='cuda:0'), tensor(18426878., device='cuda:0'), tensor(18792218., device='cuda:0'), tensor(18774330., device='cuda:0'), tensor(20196464., device='cuda:0'), tensor(19617586., device='cuda:0'), tensor(17483362., device='cuda:0'), tensor(19204476., device='cuda:0'), tensor(19153652., device='cuda:0')]

Others: [{'iter_num': 9, 'num_positive': tensor(2390, device='cuda:0'), 'num_positive_true': tensor(20211, device='cuda:0')}, {'iter_num': 9, 'num_positive': tensor(2394, device='cuda:0'), 'num_positive_true': tensor(20211, device='cuda:0')}, {'iter_num': 9, 'num_positive': tensor(2388, device='cuda:0'), 'num_positive_true': tensor(20211, device='cuda:0')}, {'iter_num': 9, 'num_positive': tensor(2396, device='cuda:0'), 'num_positive_true': tensor(20211, device='cuda:0')}, {'iter_num': 9, 'num_positive': tensor(2400, device='cuda:0'), 'num_positive_true': tensor(20211, device='cuda:0')}, {'iter_num': 9, 'num_positive': tensor(2404, device='cuda:0'), 'num_positive_true': tensor(20211, device='cuda:0')}, {'iter_num': 9, 'num_positive': tensor(2401, device='cuda:0'), 'num_positive_true': tensor(20211, device='cuda:0')}, {'iter_num': 9, 'num_positive': tensor(2408, device='cuda:0'), 'num_positive_true': tensor(20211, device='cuda:0')}, {'iter_num': 7, 'num_positive': tensor(2403, device='cuda:0'), 'num_positive_true': tensor(20211, device='cuda:0')}, {'iter_num': 9, 'num_positive': tensor(2383, device='cuda:0'), 'num_positive_true': tensor(20211, device='cuda:0')}]

Compressed training loss: [tensor(40768.0273, device='cuda:0'), tensor(40797.4609, device='cuda:0'), tensor(40828.0156, device='cuda:0'), tensor(40829.1602, device='cuda:0'), tensor(40849.9805, device='cuda:0'), tensor(40767.2188, device='cuda:0'), tensor(40818.2422, device='cuda:0'), tensor(40922.4141, device='cuda:0'), tensor(40922.6055, device='cuda:0'), tensor(40898.3906, device='cuda:0')]

Training loss: 0

Prediction time: [datetime.timedelta(seconds=1, microseconds=35640), datetime.timedelta(seconds=1, microseconds=26646), datetime.timedelta(seconds=1, microseconds=17684), datetime.timedelta(seconds=1, microseconds=42578), datetime.timedelta(seconds=1, microseconds=14696), datetime.timedelta(seconds=1, microseconds=27627), datetime.timedelta(seconds=1, microseconds=23657), datetime.timedelta(microseconds=999760), datetime.timedelta(microseconds=987811), datetime.timedelta(seconds=1, microseconds=16687)]

Phi time: [datetime.timedelta(microseconds=239993), datetime.timedelta(microseconds=250936), datetime.timedelta(microseconds=232016), datetime.timedelta(microseconds=258903), datetime.timedelta(microseconds=255915), datetime.timedelta(microseconds=237009), datetime.timedelta(microseconds=257907), datetime.timedelta(microseconds=254919), datetime.timedelta(microseconds=234008), datetime.timedelta(microseconds=235999)]

