Skip to content

Python program that recognizes 28 x 28 handwritten digits from the MNIST dataset using ReLU, softmax, and back/forward propogation all from scratch.

Notifications You must be signed in to change notification settings

lushkiwi/ML-MNIST-from-scratch

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

2 Commits
 
 
 
 
 
 
 
 

Repository files navigation

This is a from-scratch, homegrown MNIST digit-recognition AI. Accuracy depends on how much training is ran. Completely plug and play, just download the code + dataset and go train the model. On testing, it will show the handwritten digit and print the expected value with confidence/accuracy in the console. Was made without any machine learning algorithms, math notes are attatched.

https://medium.com/p/71084c6b8f91 (article explaining absolutely everything involved)

About

Python program that recognizes 28 x 28 handwritten digits from the MNIST dataset using ReLU, softmax, and back/forward propogation all from scratch.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages