Skip to content

redst4r/stacked_denoising_autoencoders

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

6 Commits
 
 
 
 
 
 

Repository files navigation

stacked_denoising_autoencoders

Trying the reproduce the results from (http://jmlr.org/papers/volume11/vincent10a/vincent10a.pdf)

Stacked Denoising Autoencoders: Learning Useful Representations in a Deep Network with a Local Denoising Criterion

Requirements:

  • lasagne
  • nolearn

The big issue for me is that I am not able to reproduce the filters the authors obtain:

  • They get blob/edge detectors
  • I get some random crap

Turns out that sample size is the issue: Going from 50000 to 500000 results in the desired filters.

See also this thread

About

trying to redo the Vincent2010 paper

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published