Special test with strict convergence condition

Precision: [tensor(0.7270, device='cuda:0'), tensor(0.7327, device='cuda:0'), tensor(0.7288, device='cuda:0'), tensor(0.7259, device='cuda:0'), tensor(0.7324, device='cuda:0'), tensor(0.7401, device='cuda:0'), tensor(0.7390, device='cuda:0'), tensor(0.7380, device='cuda:0'), tensor(0.7331, device='cuda:0'), tensor(0.7405, device='cuda:0')]

Output distance: [tensor(5.0420, device='cuda:0'), tensor(5.0402, device='cuda:0'), tensor(5.0407, device='cuda:0'), tensor(5.0459, device='cuda:0'), tensor(5.0402, device='cuda:0'), tensor(5.0297, device='cuda:0'), tensor(5.0326, device='cuda:0'), tensor(5.0381, device='cuda:0'), tensor(5.0360, device='cuda:0'), tensor(5.0273, device='cuda:0')]

Prediction loss: [tensor(16704821., device='cuda:0'), tensor(19299778., device='cuda:0'), tensor(19486180., device='cuda:0'), tensor(20493612., device='cuda:0'), tensor(18517038., device='cuda:0'), tensor(18851104., device='cuda:0'), tensor(17918066., device='cuda:0'), tensor(19261392., device='cuda:0'), tensor(18095252., device='cuda:0'), tensor(17988630., device='cuda:0')]

Others: [{'iter_num': 11, 'num_positive': tensor(2216, device='cuda:0'), 'num_positive_true': tensor(20211, device='cuda:0')}, {'iter_num': 7, 'num_positive': tensor(2177, device='cuda:0'), 'num_positive_true': tensor(20211, device='cuda:0')}, {'iter_num': 9, 'num_positive': tensor(2209, device='cuda:0'), 'num_positive_true': tensor(20211, device='cuda:0')}, {'iter_num': 11, 'num_positive': tensor(2193, device='cuda:0'), 'num_positive_true': tensor(20211, device='cuda:0')}, {'iter_num': 9, 'num_positive': tensor(2179, device='cuda:0'), 'num_positive_true': tensor(20211, device='cuda:0')}, {'iter_num': 9, 'num_positive': tensor(2193, device='cuda:0'), 'num_positive_true': tensor(20211, device='cuda:0')}, {'iter_num': 9, 'num_positive': tensor(2180, device='cuda:0'), 'num_positive_true': tensor(20211, device='cuda:0')}, {'iter_num': 7, 'num_positive': tensor(2145, device='cuda:0'), 'num_positive_true': tensor(20211, device='cuda:0')}, {'iter_num': 9, 'num_positive': tensor(2207, device='cuda:0'), 'num_positive_true': tensor(20211, device='cuda:0')}, {'iter_num': 7, 'num_positive': tensor(2208, device='cuda:0'), 'num_positive_true': tensor(20211, device='cuda:0')}]

Compressed training loss: [tensor(40817.5156, device='cuda:0'), tensor(40923.9805, device='cuda:0'), tensor(40908.0859, device='cuda:0'), tensor(40863.6250, device='cuda:0'), tensor(40728.8281, device='cuda:0'), tensor(40763.3164, device='cuda:0'), tensor(40811.1250, device='cuda:0'), tensor(40813.3906, device='cuda:0'), tensor(40898.0586, device='cuda:0'), tensor(40740.9727, device='cuda:0')]

Training loss: 0

Prediction time: [datetime.timedelta(seconds=1, microseconds=33661), datetime.timedelta(seconds=1, microseconds=22706), datetime.timedelta(seconds=1, microseconds=38679), datetime.timedelta(seconds=1, microseconds=25694), datetime.timedelta(seconds=1, microseconds=17727), datetime.timedelta(seconds=1, microseconds=29677), datetime.timedelta(microseconds=985860), datetime.timedelta(seconds=1, microseconds=20715), datetime.timedelta(seconds=1, microseconds=28680), datetime.timedelta(seconds=1, microseconds=2790)]

Phi time: [datetime.timedelta(microseconds=238996), datetime.timedelta(microseconds=267876), datetime.timedelta(microseconds=236009), datetime.timedelta(microseconds=238001), datetime.timedelta(microseconds=235014), datetime.timedelta(microseconds=253935), datetime.timedelta(microseconds=257919), datetime.timedelta(microseconds=233021), datetime.timedelta(microseconds=238998), datetime.timedelta(microseconds=234018)]

