Awesome
Adaptive Adapter Routing for Long-Tailed Class-Incremental Learning
<div align="center"> <div> <a href='http://www.lamda.nju.edu.cn/qizh' target='_blank'>Zhi-Hong Qi</a>  <a href='http://www.lamda.nju.edu.cn/zhoudw' target='_blank'>Da-Wei Zhou</a>  <a>Yiran Yao</a>  <a href='http://www.lamda.nju.edu.cn/yehj' target='_blank'>Han-Jia Ye</a>  <a href='http://www.lamda.nju.edu.cn/zhandc' target='_blank'>De-Chuan Zhan</a> </div> <div> School of Artificial Intelligence, State Key Laboratory for Novel Software Technology, Nanjing University  </div> </div> <div align="center"> </div>The code repository for "Adaptive Adapter Routing for Long-Tailed Class-Incremental Learning" (MJL 2024) in PyTorch.
<!-- If you use any content of this repo for your work, please cite the following bib entry: @article{zhou2024revisiting, author = {Zhou, Da-Wei and Cai, Zi-Wen and Ye, Han-Jia and Zhan, De-Chuan and Liu, Ziwei}, title = {Revisiting Class-Incremental Learning with Pre-Trained Models: Generalizability and Adaptivity are All You Need}, journal = {International Journal of Computer Vision}, year = {2024} } -->🔧 Requirements
Environment
Dataset
We provide the processed datasets as follows:
-
CIFAR100: will be automatically downloaded by the code.
-
ObjectNet: Onedrive: link You can also refer to the filelist and processing code if the file is too large to download.
These subsets are sampled from the original datasets. Please note that I do not have the right to distribute these datasets. If the distribution violates the license, I shall provide the filenames instead.
You need to modify the path of the datasets in ./utils/data.py according to your own path.
💡 Running scripts
To prepare your JSON files, refer to the settings in the exps folder and run the following command. The results can be found in the logs folder.
python main.py --config ./exps/[configname].json
🎈 Acknowledgement
This repo is based on LAMDA-PILOT, CIL_Survey and PyCIL.
💠Correspondence
If you have any questions, please contact me via email or open an issue.