This project implements an AI-powered handwritten digit recognition system using the MNIST dataset. The model is trained with deep learning techniques to classify handwritten digits (0–9) with high accuracy.
- MNIST dataset preprocessing
- Neural network model training
- Real-time digit prediction
- Model evaluation and accuracy visualization
- Training loss and performance analysis
- Clean modular project structure
- Python
- TensorFlow / PyTorch
- NumPy
- Matplotlib
- Jupyter Notebook
The purpose of this project is to practice and demonstrate:
- Computer Vision fundamentals
- Deep Learning model development
- Data preprocessing and training workflows
- AI model evaluation and visualization
- CNN-based architecture
- Web deployment with Flask/React
- Real-time canvas digit input
- Model comparison experiments