https://github.com/bluebottlewize/grahyam

Malayalam Handwriting Recognition Model
https://github.com/bluebottlewize/grahyam

Last synced: 23 days ago
JSON representation

Malayalam Handwriting Recognition Model

• Host: GitHub
• URL: https://github.com/bluebottlewize/grahyam
• Owner: bluebottlewize
• Created: 2024-10-28T08:31:31.000Z (2 months ago)
• Default Branch: main
• Last Pushed: 2024-12-06T19:42:32.000Z (about 1 month ago)
• Last Synced: 2024-12-06T20:47:16.972Z (about 1 month ago)
• Language: Jupyter Notebook
• Size: 12.4 MB
• Stars: 0
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# grahyam

A simple Malayalam Handwriting Recognition Model built from scratch.   

# This is a work in progress!

This is very much a work in progress. We need much more handwritten data to support more letters and stuff.    

Feel free to contribute using the [Lipi android app](https://github.com/bluebottlewize/lipi).

## How It Works

The lstm model take a sequence of touch point coordinates and spits out the corresponding Malayalam character.

## Dataset

You can find the touch point dataset in the `train` and `test` folder.