https://github.com/tim-learn/BA3US
code for our ECCV 2020 paper "A Balanced and Uncertainty-aware Approach for Partial Domain Adaptation"
https://github.com/tim-learn/BA3US
domain-adaptation eccv-2020 transfer-learning
Last synced: 5 months ago
JSON representation
code for our ECCV 2020 paper "A Balanced and Uncertainty-aware Approach for Partial Domain Adaptation"
- Host: GitHub
- URL: https://github.com/tim-learn/BA3US
- Owner: tim-learn
- License: mit
- Created: 2020-07-29T12:22:57.000Z (about 5 years ago)
- Default Branch: master
- Last Pushed: 2023-05-23T13:34:57.000Z (over 2 years ago)
- Last Synced: 2024-11-15T06:32:05.249Z (11 months ago)
- Topics: domain-adaptation, eccv-2020, transfer-learning
- Language: Python
- Homepage:
- Size: 4.62 MB
- Stars: 42
- Watchers: 3
- Forks: 11
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
Code for our ECCV (2020) paper [**A Balanced and Uncertainty-aware Approach for Partial Domain Adaptation**](https://arxiv.org/abs/2003.02541).
### Prerequisites:
- python == 3.6.8
- pytorch ==1.1.0
- torchvision == 0.3.0
- numpy, scipy, PIL, argparse, tqdm
### Dataset:
- Please manually download the datasets [Office](https://drive.google.com/file/d/0B4IapRTv9pJ1WGZVd1VDMmhwdlE/view), [Office-Home](https://drive.google.com/file/d/0B81rNlvomiwed0V1YUxQdC1uOTg/view), [ImageNet-Caltech](http://www.vision.caltech.edu/Image_Datasets/Caltech101/101_ObjectCategories.tar.gz) from the official websites, and modify the path of images in each '.txt' under the folder './data/'.
- We adopt the same data protocol as [PADA](https://github.com/thuml/PADA/tree/master/pytorch/data).
### Training:
1. ##### Partial Domain Adaptation (PDA) on the Office-Home dataset [Art(s=0) -> Clipart(t=1)]
```python
python run_partial.py --s 0 --t 1 --dset office_home --net ResNet50 --cot_weight 1. --output run1 --gpu_id 0
```
2. ##### Partial Domain Adaptation (PDA) on the Office dataset [Amazon(s=0) -> DSLR(t=1)]
```python
python run_partial.py --s 0 --t 1 --dset office --net ResNet50 --cot_weight 5. --output run1 --gpu_id 0
python run_partial.py --s 0 --t 1 --dset office --net VGG16 --cot_weight 5. --output run1 --gpu_id 0
```
3. ##### Partial Domain Adaptation (PDA) on the ImageNet-Caltech dataset [ImageNet(s=0) -> Caltech(t=1)]
```python
python run_partial.py --s 0 --t 1 --dset imagenet_caltech --net ResNet50 --cot_weight 5. --output run1 --gpu_id 0
```
### Citation
If you find this code useful for your research, please cite our paper
> @inproceedings{liang2020baus,
> title={A Balanced and Uncertainty-aware Approach for Partial Domain Adaptation},
> author={Liang, Jian, and Wang, Yunbo, and Hu, Dapeng, and He, Ran and Feng, Jiashi},
> booktitle={European Conference on Computer Vision (ECCV)},
> pages={xx-xx},
> month = {August},
> year={2020}
> }
### Acknowledgement
Some parts of this project are built based on the following open-source implementation
- CDAN [https://github.com/thuml/CDAN/tree/master/pytorch](https://github.com/thuml/CDAN/tree/master/pytorch)
- COT [https://github.com/henry8527/COT](https://github.com/henry8527/COT)
### Contact
- [liangjian92@gmail.com](mailto:liangjian92@gmail.com)
- [dapeng.hu@u.nus.edu](mailto:dapeng.hu@u.nus.edu)
- [elefjia@nus.edu.sg](mailto:elefjia@nus.edu.sg)