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
Hi @ahhnchangjun ,
we never had to really use ImageNet for incremental training.
In my case, most works I see either start pretrained on ImageNet or consider small subsets (such as tiny-imagenet, mini-imagenet, imagenet-200).
It should not be hard to implement using the PyTorch's custom dataset class, the main issue I think would be having it automatically downloaded.
Hello, guys
I am using the framework well, thank you!
Any idea why there is no full ImageNet dataset implementation ?
Thanks.
The text was updated successfully, but these errors were encountered: