https://github.com/birkhoffg/counternet

This is the official repository of the paper "CounterNet: End-to-End Training of Counterfactual Aware Predictions".
https://github.com/birkhoffg/counternet

counterfactual-explanations deep-learning explainability explainable-ai interpretability interpretable-machine-learning machine-learning nbdev pytorch recourse xai

Last synced: 13 days ago
JSON representation

This is the official repository of the paper "CounterNet: End-to-End Training of Counterfactual Aware Predictions".

• Host: GitHub
• URL: https://github.com/birkhoffg/counternet
• Owner: BirkhoffG
• License: apache-2.0
• Created: 2021-06-12T09:37:57.000Z (over 3 years ago)
• Default Branch: master
• Last Pushed: 2021-09-18T19:39:33.000Z (over 3 years ago)
• Last Synced: 2023-07-26T15:32:12.506Z (over 1 year ago)
• Topics: counterfactual-explanations, deep-learning, explainability, explainable-ai, interpretability, interpretable-machine-learning, machine-learning, nbdev, pytorch, recourse, xai
• Language: Jupyter Notebook
• Homepage:
• Size: 5.12 MB
• Stars: 0
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • Contributing: CONTRIBUTING.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# CounterNet: End-to-End Training of Counterfactual Aware Predictions

This is the official repository of the paper [CounterNet: End-to-End Training of Counterfactual Aware Predictions](https://birkhoffg.github.io/files/icml21_workshop/counternet_paper.pdf). The purpose of the repository is only for research and reproduction of the paper's results. The audience should not expect to use the code directly in the deployed environemnt. 

The repository is built based on [nbdev](https://nbdev.fast.ai/). I highly recommend you to check out [nbdev](https://nbdev.fast.ai/) if you enjoy writing code with Jupyter Notebook as I do. Further, this project primarily leverages `Pytorch` and `Pytorch Lightning` for implementations of deep learning models. To install all the dependencies, you should run:

```

pip install -e .

```

Note:

- `pip install` will only install cpu-version of  `pytorch`. If you want to use GPU-version of `pytorch`, please follow [pytorch's official instruction](https://pytorch.org/get-started/locally/).

- As `Pytorch Lightning`'s API changes rapidly, it is not guaranteed that the code is compatible with other versions of Lightning (except the version that specified `settings.ini`).

## Useful commands for `nbdev`

### Build nbs to module

```

nbdev_build_lib

```

### Update nbs from module

```

nbdev_update_lib

```

### clean notebooks

```

nbdev_clean_nbs

```

### Test all modules

```

nbdev_test_nbs

```

## Start tensorboard

```

tensorboard --logdir log --bind_all

```