Handwritten Digit Recognition

A simple handwritten digit recognition project using a TensorFlow/Keras CNN model and a custom GUI for drawing digits.

Project Contents

• draw_digit_gui.py - GUI app for drawing a digit, running prediction, and displaying the result.
• digit_model.h5 - Saved trained model for digit classification.
• Handwritten Digit Recognition.ipynb - Notebook showing data loading, model training, evaluation, and visualization.
• requirements.txt - Python dependencies required to run the project.

Requirements

Install dependencies with:

pip install -r requirements.txt

Usage

Run the GUI application with:

python draw_digit_gui.py

Then draw a digit in the canvas and click the Predict button to see the model's prediction.

Notes

• The model expects a 28x28 grayscale input image.
• The GUI uses customtkinter, and the model is loaded from digit_model.h5.
• If you want to retrain or inspect the model, open Handwritten Digit Recognition.ipynb.