HandWritten-Digit Recognition

This project aims to create a digit recognition system using TensorFlow for training a convolutional neural network (CNN) model and Gradio for creating a user-friendly interface. The trained model can recognize hand-drawn digits ranging from 0 to 9 with high accuracy.

Files Included

• digit-recognition.ipynb: This notebook contains the code for training the CNN model using the MNIST dataset, saving the trained model, and evaluating its performance.
• interface.ipynb: This notebook includes the code for setting up a Gradio interface where users can draw digits, and the trained model predicts the digit.

Installation

Clone the project

  git clone https://github.com/Shreyasb1015/HandWritten-Digit-Recognition

Go to the project directory

  cd HandWritten-Digit-Recognition

Install dependencies

  pip install tensorflow gradio

Implementation Examples