Skip to content

Python Multi - Layer perceptron implementation for educational purposes, trained on the hand-written digit MNIST dataset.

Notifications You must be signed in to change notification settings

kw5t45/neural_network_from_scratch

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

10 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Neural network from scratch.

Creating my first neural network from scratch using only numpy.
This is the classic neural network which recognizes hand written digits
from 0 to 9, using the MNIST database.
The images in the CSV are 28x28 (labeled),
and our goal is to get the script to recognize the correct image.

Used database: https://www.kaggle.com/datasets/oddrationale/mnist-in-csv

  • Finding 0.01 as the optimal hyperparameter - 6 epochs:

  • Training process using 0.01 learning rate

  • Network presentation & visualization

About

Python Multi - Layer perceptron implementation for educational purposes, trained on the hand-written digit MNIST dataset.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages