Harmonizing Generalization and Personalization in Federated Prompt Learning [ICML2024]

The implementation of paper Harmonizing Generalization and Personalization in Federated Prompt Learning[ICML2024]. [paper]

How to Run

You can run federated_main.py with some specified arguments.

Data Preparation

Please follow the instructions at CoOP https://github.com/KaiyangZhou/CoOp/blob/main/DATASETS.md to prepare the following datasets: Caltech101, OxfordPets, Flowers102, Food101, DTD.

For CIFAR10 and CIFAR100 datasets, please download and unzip data under DATA/ file catalog. Or simply run experiments with CIFAR10/CIFAR100 dataset, the program will download data automatically.

For DomainNet and office-caltech10 datasets, please follow the instructions of Dataset described here.

Training

--root takes as input a path to dataset.

--config-file means which config file to use.

You can select variables like shots, users by changing cfg or you can change every arguments you like in scripts.

Running example

bash scripts/plt_few_shot.sh

Citation

If you find our work useful in your research, please consider citing:

@article{cui2024harmonizing,
  title={Harmonizing Generalization and Personalization in Federated Prompt Learning},
  author={Cui, Tianyu and Li, Hongxia and Wang, Jingya and Shi, Ye},
  journal={arXiv preprint arXiv:2405.09771},
  year={2024}
}