Welcome to the repository for our paper: "DGMamba: Domain Generalization via Generalized State Space Model."
Environment details used for the main experiments.
Environment:
Python: 3.7.13
PyTorch: 1.12.1
Torchvision: 0.13.1
CUDA: 10.2
CUDNN: 7605
NumPy: 1.21.5
PIL: 9.5.0
selective_scan: 0.0.2
pip install -r requirements.txt
Below, we provide the checkpoints of DGMamba for PACS.
You can use the following training command to train DGMamba. We provide the sample on PACS with 'Art painting' as the target domain.
CUDA_VISIBLE_DEVICES='0' CUDA_LAUNCH_BLOCKING=1 python -u -m torch.distributed.launch --nproc_per_node=1 \
--master_port 11773 main.py --cfg ./configs/vssm_tiny_224_0220.yaml --data-path your_data_path --lr 3e-4\
--algorithm DGMamba --output ./train_output --dataset PACS --test_envs 0 --pretrained pretrained_file
This project is based on VMamba (paper, code). We thank their authors for making the source code publically available.
If you find DGMamba useful in your research, please consider citing:
@inproceedings{long2024dgmamba,
title={Dgmamba: Domain generalization via generalized state space model},
author={Long, Shaocong and Zhou, Qianyu and Li, Xiangtai and Lu, Xuequan and Ying, Chenhao and Luo, Yuan and Ma, Lizhuang and Yan, Shuicheng},
booktitle={Proceedings of the ACM International Conference on Multimedia},
year={2024}
}
This project is released under the Apache License 2.0, while some specific features in this repository are with other licenses. Please refer to LICENSES.md for the careful check, if you are using our code for commercial matters.