This repo trained Chinese Traditional Medicine Classification by VIT. The dataset contains 267k images with 163 class. I random split %5 for val and %5 for test. VIT get top1 acc 0.9854 on test data set.
If you want to use the model and code, please refer to https://github.com/yaoxiaoyuan/mimix.
@misc{mimix,
title={mimix},
author={Xiaoyuan Yao},
year={2021}
}