search

hyperconnect / LADE

This repository contains code for the paper "Disentangling Label Distribution for Long-tailed Visual Recognition", published at CVPR' 2021
https://arxiv.org/abs/2012.00321
Other
96 stars 11 forks source link

How to use your classifier for Causal-Norm?? #6

Closed edwardcho closed 3 years ago

edwardcho commented 3 years ago

Hello Sir,

I could solve my issues for LADE and balanced-softmax at your previous response.

But, I couldn't solve my issues completely.

I can understand your comment for LADE and balanced-softmax. (Just replacing loss function, I can try it. I can understand) Thanks...

But, still I didn't solve Causal-Norm in your code. Should I do create new network using feature-extraction layers and your classifier for other networks ??

Thanks, Edward Cho.

edwardcho commented 3 years ago

Hello Sir,

I saw causal.yaml in config/ImageNet_LT. Honesty, I don't know how to apply in other network (ex: Inception_v1).

In causal.yaml, I found 'apply_ipca' and 'num_components'. How to use it other network ??

And, for apply to Causal_norm, Should I use 'apply_ipa' & 'num_components' and 'CausalNormClassifier.py' in other network??

Thanks, Edward Cho.

juice500ml commented 3 years ago

Hi! Please check Causal norm's implementation for more details: https://github.com/KaihuaTang/Long-Tailed-Recognition.pytorch