Skip to content

an implementation of L0 regularization with PyTorch

Notifications You must be signed in to change notification settings

tyuw437ry78/l0.pytorch

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

3 Commits
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

l0.pytorch

An implementation of Learning Sparse Neural Networks through L 0 Regularization by C. Louizos, M. Welling and D.P. Kingma.

Contents

Requirements

  • PyTorch 0.3/ torchvision 0.2
  • tensorboard-pytorch
  • tqdm

Results

LeNet5

Not yet strictly measure how sparse the L0 regularized model is, but show histograms of the first convolutional layers' weights.

  • with L0 normalization

  • without L0 normalization

ToDo

  • Regularization for biases (currently only weights are regularized).
  • More complex architectures with L0 Norm.

About

an implementation of L0 regularization with PyTorch

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Jupyter Notebook 90.9%
  • Python 9.1%