Skip to content

99-WSJ/Wavelet_loss

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

22 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Wavelet_loss

Wavelet_loss Function

Implementation of Wavelet Loss Function for Auto-Encoder (https://ieeexplore.ieee.org/document/9351990) in PyTorch

image

Requirements

This script requires:

  • pytorch
  • pytorch_wavelets
  • torchvision
  • PIL

If you don't already have pytorch or torchvision please have a look at https://pytorch.org/ as the installation command may vary depending on your OS and your version of CUDA.

You can install all other dependencies with pip by running terminal and get the result images by wavelet transform.

Just run

python test.py 

wavelet transform at every level (eg. 128*128).

img_grid1_2

img_grid1_1

img_grid1_0

And by changing weight in every level, you get some different results.

image

Take a try for Autoencoder's training.

image

Reference

pytorch_wavelets : https://github.com/fbcotter/pytorch_wavelets

LICENSE

This project is under MIT license.

About

Wavelet_loss Function

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors

Languages