Skip to content

A repo for toy examples to test uncertainties estimation of neural networks

Notifications You must be signed in to change notification settings

cdebeunne/uncertainties_CNN

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

23 Commits
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Uncertainties Neural Networks

A notebook that implements a few toy neural networks in order to test uncertainties estimation.

  • A 1D regression with the CO2 dataset
  • An image classifier for MNIST's data
  • A custom pose estimator for 2D squares, inpired by Yann Labbe's 6D object tracker cosypose https://github.com/ylabbe/cosypose

Two methods are explored in the associated notebook:

The code of ADF's function (contrib folder) is provided by Mattia Segù and available @ https://github.com/mattiasegu/uncertainty_estimation_deep_learning

About

A repo for toy examples to test uncertainties estimation of neural networks

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published