Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/BIT-DA/Domain-Oriented-Transformer
https://github.com/BIT-DA/Domain-Oriented-Transformer
Last synced: 7 days ago
JSON representation
- Host: GitHub
- URL: https://github.com/BIT-DA/Domain-Oriented-Transformer
- Owner: BIT-DA
- License: mit
- Created: 2022-07-12T12:54:02.000Z (over 2 years ago)
- Default Branch: main
- Last Pushed: 2022-11-24T12:07:30.000Z (almost 2 years ago)
- Last Synced: 2024-08-02T15:26:04.391Z (3 months ago)
- Language: Python
- Size: 4.11 MB
- Stars: 5
- Watchers: 0
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
---
# Making the Best of Both Worlds: A Domain-Oriented Transformer for Unsupervised Domain Adaptation
Wenxuan Ma, Jinming Zhang, [Shuang Li](https://shuangli.xyz), , [Chi Harold Liu](https://scholar.google.com/citations?user=3IgFTEkAAAAJ&hl=en), [Yulin Wang](https://www.rainforest-wang.cool/), and Wei Li
[](https://arxiv.org/abs/2208.01195)
Official implementation of our ACM MM 2022 paper "Making the Best of Both Worlds: A Domain-Oriented Transformer for Unsupervised Domain Adaptation" (DOT).
## Framework Overview
In this work, we propose a new UDA paradigm that simultaneously learns multiple feature embedding spaces with their own different specification (denoted as **domain-oriented feature spaces**). The comparison between our new paradigm and the classical ones is shown in the figure.
## Datasets Preparation
### Office-Home
Office-home dataset can be download from [here](https://www.hemanthdv.org/officeHomeDataset.html).
### VisDA-2017
VisDA-2017 dataset can be download from [here](http://ai.bu.edu/visda-2017/).
### DomainNet
DomainNet dataset can be download from [here](http://ai.bu.edu/M3SDA/).
When finish downloading, please modify the corresponding data lists in "./data/" to match your data root.
## Training DOT for Unsupervised Domain Adaptation
### Source Model
First, download the ImageNet-1k pretrained [vit_small](https://dl.fbaipublicfiles.com/deit/deit_small_distilled_patch16_224-649709d9.pth) and [vit_base](https://dl.fbaipublicfiles.com/deit/deit_base_distilled_patch16_224-df68dfff.pth) model, and put them under the "./pretrained/" directory.
Next, a source model is trained to generate initial pseudo-labels for DOT training. The source model can be downloaded [here](https://github.com/BIT-DA/Domain-Oriented-Transformer/releases).
Or you can run the following command to train the source model by yourself:
```python
# Office-Home [0,1,2,3]->[Ar,Cl,Pr,Rw]
python train_sourceonly.py --dset home --s 0 --model vit_small --output_src [YOUR_SOURCE_MODEL_ROOT] --mixup-active --save --seed 2022 --gpu_id 0
# VisDA2017 [0,1]->[Synthetic,Real]
python train_sourceonly.py --dset home --s 0 --model vit_small --output_src [YOUR_SOURCE_MODEL_ROOT] --mixup-active --save --seed 2022 --gpu_id 0
```
### Training DOT
Run the following command to train a DOT model.
```python
# Office-Home [0,1,2,3]->[Ar,Cl,Pr,Rw]
python train_DOT.py --dset home --s 0 --t 1 --model dot_small --output_dir output --src_model_path [YOUR_SOURCE_MODEL_ROOT] --seed 2022 --gpu_id 0
# VisDA2017 [0,1]->[Synthetic,Real]
python train_DOT.py --dset visda2017 --s 0 --t 1 --model dot_small --output_dir output --src_model_path [YOUR_SOURCE_MODEL_ROOT] iter_per_epoch 1000 --lr 1e-3 --seed 2022 --gpu_id 0
```
## Acknowledgement
This repository borrows codes from the following repos. Many thanks to the authors for their great work.
timm-library: https://github.com/rwightman/pytorch-image-models
SHOT: https://github.com/tim-learn/SHOT
SupContrast: https://github.com/HobbitLong/SupContrast
## Citation
If you find this project useful, please consider citing:
```
@inproceedings{ma2022making,
title={Making the Best of Both Worlds: A Domain-Oriented Transformer for Unsupervised Domain Adaptation},
author={Ma, Wenxuan and Zhang, Jinming and Li, Shuang and Liu, Chi Harold and Wang, Yulin and Li, Wei},
booktitle={Proceedings of the 30th ACM International Conference on Multimedia},
pages={5620--5629},
year={2022}
}
```
## Contact
If you have any questions about our code, feel free to contact us or describe your problem in [Issues](https://github.com/BIT-DA/SePiCo/issues/new).
Email address: [email protected]; [email protected].