Skip to content
Snippets Groups Projects
Select Git revision
  • master default protected
1 result
Compare
hoffma31's avatar
renamed module
Lara Hoffmann authored
ab836d2b
History
hoffma31's avatar ab836d2b
History
Name Last commit Last update
bdeepens
LICENSE
README.md
notebook.ipynb
README.md

BayesianDeepEnsembles

This repository provides the python code to reproduce the results from the paper "Deep Ensembles from a Bayesian Perspective" [1], and additional examples. The paper introduces a simple extension to the state-of-the-art method deep ensembles, resulting in an improved uncertainty quantification. This can be achieved by an additional post-processing step, which is implemented in this repository.

References

[1] L. Hoffmann and C. Elster, Deep Ensembles from a Bayesian Perspective, arXiv preprint, 2021. [https://arxiv.org/abs/2105.13283]

License

copyright: Lara Hoffmann (PTB) 2021

This software is licensed under the BSD-like license:

Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met:

  1. Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer.
  2. Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the following disclaimer in the documentation and/or other materials provided with the distribution.

DISCLAIMER

This software was developed at Physikalisch-Technische Bundesanstalt (PTB). The software is made available "as is" free of cost. PTB assumes no responsibility whatsoever for its use by other parties, and makes no guarantees, expressed or implied, about its quality, reliability, safety, suitability or any other characteristic. In no event will PTB be liable for any direct, indirect or consequential damage arising in connection

Using this software in publications requires citing the following paper

L. Hoffmann and C. Elster, Deep Ensembles from a Bayesian Perspective, arXiv preprint, 2021. [https://arxiv.org/abs/2105.13283]

Usage

The usage is explained in the file notebook.ipynb.

Note that the implementation currently only works properly for the output dimension p_y = 1.

The code is running on python 3.6 and uses the following packages

  • matplotlib
  • numpy
  • pytorch
  • random
  • copy

Contact

In case of problems or questions please contact lara.hoffmann@ptb.de.