Skip to content

Helper functions for estimating uncertainty with neural networks

License

Notifications You must be signed in to change notification settings

ae137/nn_uncertainty_estimation

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

7 Commits
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Neural network uncertainty estimation

Content of package

This package contains a collection of additions to Tensorflow.

Currently, it includes

  • losses.py:

    • Implementation of Negative Log Likelihood loss for regression tasks
  • metrics.py:

    • Extension of MSE suitable for a networks that predicts means and variances

Building and installing the package

Run

  1. pip install -r requirements.txt for installing the build requirements
  2. python -m build for building the package
  3. pip install dist/nn_uncertainty_estimation-0.0.1-py3-none-any.whl

TODO: Include step for running the tests automatically.

About

Helper functions for estimating uncertainty with neural networks

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages