search

TACJu / TransFG

This is the official PyTorch implementation of the paper "TransFG: A Transformer Architecture for Fine-grained Recognition" (Ju He, Jie-Neng Chen, Shuai Liu, Adam Kortylewski, Cheng Yang, Yutong Bai, Changhu Wang, Alan Yuille).
MIT License
382 stars 88 forks source link

Normalization parameters #20

Open BinahHu opened 2 years ago

BinahHu commented 2 years ago

Thank you for your excellent work! I am just wondering why all the datasets use the same normalization parameters (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) and they are the same as ImageNet?

Kolaye commented 2 years ago

Yes，I have the same question！Why not recalculate the mean and std with the new dataset！

ngthanhtin commented 11 months ago

Hi @BinahHu , @Kolaye , because they loaded pretrained weights trained on ImageNet-21k, so they used the mean and std of the ImageNet dataset.