Digit Recognizer is a web-based application which can recognise handwritten digits and classify them. This application is created for the purpose of learning to build a Machine Learning model and deploying it as an app. So, Handwritten Digit Recognition model is based of a Deep Learning architecture and built on TensorFlow framework can recognise digits and classify them into classes. The web application is built primarily on Node.js framework.
Web-app link: https://digit-recognizer-web.herokuapp.com
- Node.js
- Express.js
- Bootstrap
- body-parser
- jQuery
- Ajax
- npm
- Tensorflow.js
- Tensorflow
- Keras
- npm
- git
- python
- node.js
- To learn more and to train yourself the Deep Learning model you can refer the interactive python notebook here.
- To run the application locally on your machine, follow these instructions.
- Clone the repository into your machine.
- Navigate to the project directory in your terminal or command prompt
- Execute
npm install
command. - Execute
npm run
command. Now the app will be opened on your browser running in a localhost.
The application is hosted in Heroku- a free cloud platform. The DL model has been developed and trained in Google Colab- free jupyter notebook environment for developing and training ML models in cloud.
This project has been licensed under MIT License. You can check the License file to view the license