Skip to content

Latest commit

 

History

History
11 lines (8 loc) · 722 Bytes

README.md

File metadata and controls

11 lines (8 loc) · 722 Bytes

Hopfield Networks

Implementation and Analysis of Hopfield Networks in Python.

Hopfield Networks are based on the idea of "associate" (content-addressable) memory and often compared to human's associative memory. (All information was taken from Wikipedia, The Free Encyclopedia; https://en.wikipedia.org/wiki/Hopfield_network)

Here I trained and tested Hopfield Networks on handwritten digits and on digits from MNIST (uniformly distributed, randomly selected subset). Different training sets yielded different results.

List of "Jupyter nbviewer" links for Jupyter Notebooks in this folder:

  1. Hopfield Network: https://nbviewer.jupyter.org/github/podolskyDavid/hopfield-network/blob/master/Hopfield%20Net.ipynb