You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Thanks for your GREAT work!! And the released code really helps a lot!!
But when I tried to replicate the experimental results on CIFAR10, I failed and only got 0.35 accuracy.
It might because of the inappropriate hyper-parameter setting, or my misunderstanding on other experimental details on CIFAR10.
I've tried different beta values (0.1, 1,5, 10) and d values (5, 10). And for the feature extraction on CIFAR10, here's my implementation:
After adding a pre-processing step, the ACC can reach 0.457 for now. But there is still a gap between 0.627 which is reported in the paper. Could you please leave more details? It would be very helpful!
Thanks for your GREAT work!! And the released code really helps a lot!!
But when I tried to replicate the experimental results on CIFAR10, I failed and only got 0.35 accuracy.
It might because of the inappropriate hyper-parameter setting, or my misunderstanding on other experimental details on CIFAR10.
I've tried different beta values (0.1, 1,5, 10) and d values (5, 10). And for the feature extraction on CIFAR10, here's my implementation:
Could you please give me some advice on the replication of CIFAR10? It would be extremely helpful!! Thanks a lot!!
The text was updated successfully, but these errors were encountered: