Skip to content

Bangla character recognizer web app based on fine tuned CNN model.

Notifications You must be signed in to change notification settings

SaihanTaki/BHWCR

Repository files navigation

Bangla Handwritten Character Recognizer

This is a web app that can recognize bangla handwritten characters. Recognition task is done by using a fine tuned convolutional neural network model.

Click here to see live demo
This web app may take upto a minute to load. Pease waite a little bit.


Table of Contents


Technologies

  • Tensorflow
  • Keras
  • Python
  • Flask
  • HTML
  • CSS
  • Javascript

Screen Capture


Model Architecture


How to run this app

  • Create a project directory in your local machine
  • cd to the project directory
  • Run the command in your shell to clone the repo or simply download the zip file
$ git clone https://github.com/SaihanTaki/Bangla-Character-Recognition.git
$ cd Bangla-Character-Recognition
  • Create a virtual environment
  • Activate the virtual environment
  • Install the dependency by running the following command
$ pip install -r requirements.txt
  • Run the application using the following command
Flask run 

Or

gunicorn --bind 0.0.0.0:5000 app:app

N.B. Flask run will start the application in development mode.


Dataset Source

Ekush Dataset

About

Bangla character recognizer web app based on fine tuned CNN model.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published