maniospas/OT-ICNN-bias

The original project aimed at learning the optimal transport map via input convex neural neworks. We want to exploit this technique to discover bias in image data.

Optimal Transport for Bias Detection in Images

This project aims at exploiting Optimal Transport techniques for the purpose of bias detection in image data.

This work is inspired by the paper optimal transport mapping via input convex neural neworks

The repository is forked from https://github.com/AmirTag/OT-ICNN

Get Started

To use our code and reproduce our results please follow the following steps.

1. Once you have forked and cloned the repository, you can create a conda enviroment

conda env create -f environment.yml

conda activate ot-bias

2. install the package locally

pip install -e .

3. Download CelebA data and place it in the folder ./OT-ICNN-bias/data/celeba or open a python console and run:

>>> import torchvision
>>> torchvision.datasets.CelebA("./data", download=True)

Make sure that the structure of the directory data is the following:

data
└── celeba
    ├── img_align_celeba
    │   ├── 000001.jpg
    │   ├── 000002.jpg
    .   .
    .   .
    .   .
    │   └── 202599.jpg
    ├── list_attr_celeba.csv
    ├── list_bbox_celeba.csv
    ├── list_eval_partition.csv
    └── list_landmarks_align_celeba.csv

4. Extract features from CelebA's images by running:

cd scripts
python ./Feature_extraction.py

Note that we use Hydra configuration manager, you can override some of the configurations by running:

python ./Feature_extraction.py features="resnet50"

5. Select suitable splits of CelebA data by running:

python ./Select_data_experiment1.py

6. Train the ICCNs. The code saves results and checkpoints in the folder ./results

python ./celeba_training.py

Again you can override the configurations, for example

python ./celeba_training.py training.epochs=50 training.optimizer="RMSProp" settings.no_cuda=True

7. Hydra stores the configuration used for each run into folders named ./results/training/DATASET_NAME/YYYY-MM-DD/HH-MM-SS/.hydra You can test the goodness of the training with your configuration by running pytest. You need to specify the epoch checkpoint you want to test as well.

cd ../tests

pytest --config='YYYY-MM-DD/HH-MM-SS/' --dataset='celeba' --epoch_to_test=40 --verbose

Acknowledgments

* https://github.com/AmirTag/OT-ICNN

References

@Misc{Yadan2019Hydra,
  author =       {Omry Yadan},
  title =        {Hydra - A framework for elegantly configuring complex applications},
  howpublished = {Github},
  year =         {2019},
  url =          {https://github.com/facebookresearch/hydra}
}