This project implements a Convolutional Neural Network (CNN) to classify handwritten digits from the MNIST dataset. The MNIST dataset consists of 60,000 training images and 10,000 test images, each representing digits from 0 to 9 in 28x28 grayscale format. The CNN is designed to achieve high accuracy in digit classification through deep learning techniques.

• Preprocessing: Normalization and reshaping of MNIST images.
• CNN Architecture: Multi-layer convolutional network with pooling and dropout.
• Training & Validation: Model trained on MNIST dataset with performance evaluation.
• Performance Metrics: Accuracy, loss, and confusion matrix visualization.
• Easy to Use: Simple implementation with TensorFlow/Keras.

Ensure you have the following installed:

• Python 3.x
• TensorFlow
• Keras
• NumPy
• Matplotlib
• scikit-learn

You can install the required dependencies using:

pip install tensorflow keras numpy matplotlib scikit-learn

The MNIST dataset is automatically downloaded when using TensorFlow/Keras. No manual download is required.

The CNN model consists of the following layers:

1. Conv2D - Extracts features using a 3x3 filter.
2. MaxPooling2D - Reduces spatial dimensions (2x2 pool size).
3. Conv2D - Another convolutional layer to learn deeper features.
4. MaxPooling2D - Further spatial reduction.
5. Flatten - Converts feature maps into a 1D vector.
6. Dense (Fully Connected Layer) - Final classification using Softmax activation.

• Achieves 98%+ accuracy on the MNIST test set.
• Loss and accuracy plots provided for training analysis.
• Confusion matrix for detailed performance insights.