Skip to content

Latest commit

 

History

History
19 lines (12 loc) · 1011 Bytes

README.md

File metadata and controls

19 lines (12 loc) · 1011 Bytes

Neural network for the recognition of handwritten digits implemented in MATLAB

This was made as a project for my scientific computing class.

The code is in MATLAB R2023a and implements a 3 layer neural network using the MNIST database of handwritten digits as the input training data.

The code follows the book "Neural Networks and Deep Learning", mainly the first chapter.

The MNIST database of handwritten digits is used in MATLAB format from this repository. For this code to run, the database has to be in the same folder as the code.

Neural Network Parameters

The Neural Network consist of 3 layers:

  • Input layers with 784 nodes
  • Hidden layer with 15 nodes
  • Output layer with 10 nodes

Note

The parameters can be changed in the code by changing the variables inputSize, hiddenSize, and outputSize. but this will require further changes in the code for it to run.