Skip to content

code for "Semi-supervised Domain Adaptation via Prototype-based Multi-level Learning"

Notifications You must be signed in to change notification settings

bupt-ai-cz/ProML

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

30 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Semi-supervised Domain Adaptation via Prototype-based Multi-level Learning

[Project] [Paper] [Supplementary Version]

Overview

Motivation

To avoid misunderstandings, let us elaborate further on our motivation and give a Supplementary Version.

1. Requirements

pip install -r requirements.txt

2. Data Preparation

For DomainNet, please follow MME to prepare the data. The expected dataset path pattern is like your-domainnet-data-root/domain-name/class-name/images.png.

For Office-Home, please download the resized images and extract, you will get a .pkl and a .npy file, then specify their paths in loader/office_home.py.

3. Training

python -u train.py --dataset visda --base_path ./data/txt/visda/ --data_root /root/SSDA/data/visda/ --source clipart --target sketch --num 1 --log_dir ./logs --num_classes 12 --threshold2 0.4 --T 0.05

4. Acknowledgement

The code is partly based on MME and MCL. Thank them for their great work.

5. Citation

@misc{huang2023semisupervised,
      title={Semi-supervised Domain Adaptation via Prototype-based Multi-level Learning}, 
      author={Xinyang Huang and Chuang Zhu and Wenkai Chen},
      year={2023},
      eprint={2305.02693},
      archivePrefix={arXiv},
      primaryClass={cs.CV}
}

6. Contact

Xinyang Huang ([email protected])

If you have any questions, you can contact us directly.

About

code for "Semi-supervised Domain Adaptation via Prototype-based Multi-level Learning"

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages