https://github.com/wesdoyle/mnist-paint

Vanilla JS + TensorFlow paint with MNIST hand-written digit classification
https://github.com/wesdoyle/mnist-paint

cnn keras mnist-handwriting-recognition ocr-recognition tensorflow

Last synced: 2 months ago
JSON representation

Vanilla JS + TensorFlow paint with MNIST hand-written digit classification

• Host: GitHub
• URL: https://github.com/wesdoyle/mnist-paint
• Owner: wesdoyle
• Created: 2021-09-04T15:54:52.000Z (almost 5 years ago)
• Default Branch: main
• Last Pushed: 2021-09-06T14:07:52.000Z (almost 5 years ago)
• Last Synced: 2025-06-08T00:08:55.415Z (about 1 year ago)
• Topics: cnn, keras, mnist-handwriting-recognition, ocr-recognition, tensorflow
• Language: Jupyter Notebook
• Homepage:
• Size: 13.2 MB
• Stars: 1
• Watchers: 2
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

          
# mnist paint

Example handwritten digit OCR using tensorflow.js and canvas.

The example model is a basic CNN trained using Keras on the MNIST handwritten digits dataset - see notebook for details.

To run, start an HTTP server in the root of the directory, e.g.:

```bash

python3 -m http.server 7771

```