https://github.com/sandialabs/quinn

Quantification of Uncertainties in Neural Networks
https://github.com/sandialabs/quinn

bayesian-inference markov-chain mcmc neural-network scr-2871 snl-applications uncertainty-quantification variational-inference

Last synced: 5 months ago
JSON representation

Quantification of Uncertainties in Neural Networks

• Host: GitHub
• URL: https://github.com/sandialabs/quinn
• Owner: sandialabs
• License: bsd-3-clause
• Created: 2023-01-10T15:22:10.000Z (over 2 years ago)
• Default Branch: main
• Last Pushed: 2025-03-30T03:12:30.000Z (6 months ago)
• Last Synced: 2025-04-06T22:38:35.794Z (6 months ago)
• Topics: bayesian-inference, markov-chain, mcmc, neural-network, scr-2871, snl-applications, uncertainty-quantification, variational-inference
• Language: Python
• Homepage: https://quinn.readthedocs.io
• Size: 7.5 MB
• Stars: 11
• Watchers: 3
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

          
Quantification of Uncertainties in Neural Networks (QUiNN) is a python library centered around various probabilistic wrappers over PyTorch modules in order to provide uncertainty estimation in Neural Network (NN) predictions.

# Build the library

	./build.sh 

	or 

	./setup.py build; setup.py install

# Requirements

	numpy, scipy, matplotlib, pytorch

# Examples

	examples/ex_fit.py

	examples/ex_fit_2d.py

	examples/ex_ufit.py  # where method=mcmc, ens or vi.

# Authors

	Khachik Sargsyan

	Javier Murgoitio-Esandi

 	Oscar Diaz-Ibarra

  

# Acknowledgements

	This work is supported by 

	- U.S. Department of Energy, Office of Fusion Energy Sciences (OFES) under Field Work Proposal Number 20-023149.

	- Laboratory Directed Research and Development (LDRD) program of Sandia National Laboratories.