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
384 stars 87 forks source link

About INatureList17 acc #34

Open ShihaoZhao-2019 opened 6 months ago

ShihaoZhao-2019 commented 6 months ago

I want to know how to set batch size and lr on INatureList17，Does anyone know?

jpoberhauser commented 6 months ago

You can set training batch size with: --train_batch_size 2

and lr with --learning_rate the current default is 3e-2

ShihaoZhao-2019 commented 6 months ago

@jpoberhauser Good, thank you very much