torch-uncertainty 0.1.3

A PyTorch Library for benchmarking and leveraging efficient predictive uncertainty quantification techniques.

TorchUncertainty is a package designed to help you leverage uncertainty quantification techniques and make your neural networks more reliable. It aims at including as many methods as possible, so reach out to add yours!

:construction: TorchUncertainty is in early development :construction: - expect massive changes but do reach out to us and contribute if you are interested by the project!

---

This package provides a multi-level API, including:

• ready-to-train baselines on research datasets, such as ImageNet and CIFAR
• baselines available for training on your datasets
• pretrained weights for these baselines on ImageNet and CIFAR (work in progress 🚧).
• layers available for use in your networks
• scikit-learn style post-processing methods such as Temperature Scaling

See the Reference page or the API reference for a more exhaustive list of the implemented methods, datasets, metrics, etc.

Installation

Install the desired pytorch version in your environment. Then, the package can be installed from PyPI:

pip install torch-uncertainty

If you aim to contribute (thank you!), have a look at the contribution page.

Getting Started and Documentation

Please find the documentation at torch-uncertainty.github.io.

A quickstart is available at torch-uncertainty.github.io/quickstart.

Implemented methods

Baselines

To date, the following baselines are implemented:

• Deep Ensembles
• BatchEnsemble
• Masksembles
• Packed-Ensembles (see blog post)
• Bayesian Neural Networks

Post-processing methods

To date, the following post-processing methods are implemented:

• Temperature scaling

Tutorials

Awesome Uncertainty repositories

You may find a lot of papers about modern uncertainty estimation techniques on the Awesome Uncertainty in Deep Learning.

Other References

This package also contains the official implementation of Packed-Ensembles.

If you find the corresponding models interesting, please consider citing our paper:

@inproceedings{laurent2023packed,
    title={Packed-Ensembles for Efficient Uncertainty Estimation},
    author={Laurent, Olivier and Lafage, Adrien and Tartaglione, Enzo and Daniel, Geoffrey and Martinez, Jean-Marc and Bursuc, Andrei and Franchi, Gianni},
    booktitle={ICLR},
    year={2023}
}