https://github.com/redjasm/hand-written-digits-classifier

A neural network made using python and numpy to classify images of hand written digits between 0 and 9. Trained with the MNIST dataset.
https://github.com/redjasm/hand-written-digits-classifier

Last synced: 7 days ago
JSON representation

A neural network made using python and numpy to classify images of hand written digits between 0 and 9. Trained with the MNIST dataset.

• Host: GitHub
• URL: https://github.com/redjasm/hand-written-digits-classifier
• Owner: redjasm
• Created: 2023-03-11T16:30:11.000Z (over 1 year ago)
• Default Branch: main
• Last Pushed: 2023-03-11T16:42:22.000Z (over 1 year ago)
• Last Synced: 2024-09-13T08:36:45.528Z (2 months ago)
• Language: Python
• Size: 107 KB
• Stars: 0
• Watchers: 2
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# Simple MNIST NN from scratch using numpy

![](MnistExamples.png)

Thanks to [Samson Zhang](https://www.kaggle.com/code/wwsalmon/simple-mnist-nn-from-scratch-numpy-no-tf-keras/notebook) for the amazing tutorial.

![](screenshot.png)

As you see in the screenshot above, the model is able to predict the handwritten digits with 83% accuracy. This is not bad for a model that is trained from scratch using only numpy. But we can do better.