MNIST Digit Recognition using CNN

This project builds a Convolutional Neural Network (CNN) to classify handwritten digits from the MNIST dataset.

Steps:

• Data Preprocessing
• CNN Model Building
• Model Training & Evaluation
• Accuracy & Confusion Matrix Visualization

Accuracy:

Achieved ~97% validation accuracy.

Technologies:

• Python
• TensorFlow / Keras
• Matplotlib & Seaborn
• Jupyter Notebook

Author:

Krupal Sidhdhapura - B.Tech [Computer Engineering], [G H Patel College Of Engineering and Technology]