Special test with strict convergence condition

Precision: [tensor(0.4187, device='cuda:0'), tensor(0.4206, device='cuda:0'), tensor(0.4177, device='cuda:0'), tensor(0.4277, device='cuda:0'), tensor(0.4277, device='cuda:0'), tensor(0.4222, device='cuda:0'), tensor(0.4193, device='cuda:0'), tensor(0.4214, device='cuda:0'), tensor(0.4206, device='cuda:0'), tensor(0.4180, device='cuda:0')]

Output distance: [tensor(5.4686, device='cuda:0'), tensor(5.4650, device='cuda:0'), tensor(5.4707, device='cuda:0'), tensor(5.4508, device='cuda:0'), tensor(5.4508, device='cuda:0'), tensor(5.4618, device='cuda:0'), tensor(5.4676, device='cuda:0'), tensor(5.4634, device='cuda:0'), tensor(5.4650, device='cuda:0'), tensor(5.4702, device='cuda:0')]

Prediction loss: [tensor(17523994., device='cuda:0'), tensor(19340350., device='cuda:0'), tensor(19806000., device='cuda:0'), tensor(18134580., device='cuda:0'), tensor(17738938., device='cuda:0'), tensor(18680918., device='cuda:0'), tensor(20370528., device='cuda:0'), tensor(18495748., device='cuda:0'), tensor(18654652., device='cuda:0'), tensor(17937952., device='cuda:0')]

Others: [{'iter_num': 5, 'num_positive': tensor(3809, device='cuda:0'), 'num_positive_true': tensor(20211, device='cuda:0')}, {'iter_num': 3, 'num_positive': tensor(3809, device='cuda:0'), 'num_positive_true': tensor(20211, device='cuda:0')}, {'iter_num': 5, 'num_positive': tensor(3809, device='cuda:0'), 'num_positive_true': tensor(20211, device='cuda:0')}, {'iter_num': 3, 'num_positive': tensor(3809, device='cuda:0'), 'num_positive_true': tensor(20211, device='cuda:0')}, {'iter_num': 5, 'num_positive': tensor(3809, device='cuda:0'), 'num_positive_true': tensor(20211, device='cuda:0')}, {'iter_num': 5, 'num_positive': tensor(3809, device='cuda:0'), 'num_positive_true': tensor(20211, device='cuda:0')}, {'iter_num': 5, 'num_positive': tensor(3809, device='cuda:0'), 'num_positive_true': tensor(20211, device='cuda:0')}, {'iter_num': 5, 'num_positive': tensor(3809, device='cuda:0'), 'num_positive_true': tensor(20211, device='cuda:0')}, {'iter_num': 5, 'num_positive': tensor(3809, device='cuda:0'), 'num_positive_true': tensor(20211, device='cuda:0')}, {'iter_num': 5, 'num_positive': tensor(3809, device='cuda:0'), 'num_positive_true': tensor(20211, device='cuda:0')}]

Compressed training loss: [tensor(40832.0586, device='cuda:0'), tensor(40931.2969, device='cuda:0'), tensor(40835.3711, device='cuda:0'), tensor(41064.5156, device='cuda:0'), tensor(40784.8594, device='cuda:0'), tensor(40932.1445, device='cuda:0'), tensor(40907.6992, device='cuda:0'), tensor(40898.9883, device='cuda:0'), tensor(40936.1133, device='cuda:0'), tensor(40850.7148, device='cuda:0')]

Training loss: 0

Prediction time: [datetime.timedelta(seconds=6, microseconds=330153), datetime.timedelta(seconds=3, microseconds=981115), datetime.timedelta(seconds=6, microseconds=191739), datetime.timedelta(seconds=4, microseconds=31900), datetime.timedelta(seconds=6, microseconds=100128), datetime.timedelta(seconds=6, microseconds=225595), datetime.timedelta(seconds=6, microseconds=62289), datetime.timedelta(seconds=6, microseconds=97142), datetime.timedelta(seconds=6, microseconds=58435), datetime.timedelta(seconds=6, microseconds=145202)]

Phi time: [datetime.timedelta(microseconds=237990), datetime.timedelta(microseconds=370430), datetime.timedelta(microseconds=357485), datetime.timedelta(microseconds=340556), datetime.timedelta(microseconds=367442), datetime.timedelta(microseconds=372421), datetime.timedelta(microseconds=369434), datetime.timedelta(microseconds=384346), datetime.timedelta(microseconds=380386), datetime.timedelta(microseconds=323641)]

