Skip to content

Latest commit

 

History

History
18 lines (15 loc) · 1.18 KB

README.md

File metadata and controls

18 lines (15 loc) · 1.18 KB

MNIST_NeuralNetworkV2

Simple neural network to classify and written digits. Impliments the back propagation algorithm. Version 2 of https://github.com/pateldeev/MNIST_NeuralNetwork

Data Set

http://yann.lecun.com/exdb/mnist/

Credit/Inspiration

https://youtu.be/aircAruvnKk

Running

Requirements

Requires python3. (Tested with 3.6.7). Make sure to have opencv and numpy

  • pip install opencv-python-headless
  • pip install numpy

Decompressing Data

This repository contains all the images in two compressed files (one with 60000 training images) and (one with 10000 validation images). The repo also contains two files with the labels for these images. They should be decompressed and saved as .png files. The data_loader module can accomplish this. Simply uncomment the appropriate code in main.py