Skip to content

Manual RNN in Numpy and simpler equivalent in PyTorch

Notifications You must be signed in to change notification settings

InexperiencedMe/RNNstudy

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

4 Commits
 
 
 
 
 
 

Repository files navigation

RNNstudy

Manual RNN in Numpy and simpler equivalent in PyTorch

The task solved is binary addition, bit by bit, which is a very simple problem, that's impossible to do for a simple multilayer perceptron, as it requires remembering the recent past bits to correctly process the sequence.

The whole inspiration came from Andrew Trask and his blog post, that I went through, analyzed the code line by line, updated the code to more modern standards and did the same in PyTorch to have a reference of how to use it there with an autograd engine.

About

Manual RNN in Numpy and simpler equivalent in PyTorch

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages