Skip to content

Latest commit

 

History

History
19 lines (13 loc) · 684 Bytes

README.md

File metadata and controls

19 lines (13 loc) · 684 Bytes

A-Connect Library


All the files needed to develop a library for easy deployment of A-Connect.

General Info

Algorithm description

A-Connect is a methodology that provides to the users a statistical way to train their neural networks. One of the A-Connect advantages is the mismatch mitigation due to the stochastic variations. This methodology also allows the binarization of the layer weights to reduce the memory footprint of the layer.

Table of Contents

  1. Tensorflow
  2. Pytorch

Contents description

Above you can see the name of two popular frameworks for Machine/Deep learning where in a future A-Connect will be available.