Skip to content

Latest commit

 

History

History
5 lines (3 loc) · 521 Bytes

README.md

File metadata and controls

5 lines (3 loc) · 521 Bytes

This project recreates some of the properties of a biological neural network in a cell grid (CNN). The goal is to be able to train the grid using only a global reward/punish value calculated from the output, and no backpropagation.

To install, just copy grid_animation.py and grid_rules.py into the desired folder and run grid_animation.py. Requires numpy (for the math) and pyglet (for the visualization).

200x200 grid showing neuron activations