This project demonstrates how to train a neural network to classify handwritten digits from the MNIST dataset using PyTorch.

The MNIST dataset contains:

• 60,000 training images

  • split into 50,000 for training, 10,000 for validation

• 10,000 test images Each image is 28×28 grayscale and represents digits 0–9.

├── MNIST.ipynb
├── models/
├── images/
└── README.md

git clone https://github.com/8Whoknow3/MNIST.git
cd mnist-classifier

Make sure you have Python 3.7+ and install the required libraries:

pip install torch torchvision matplotlib scikit-learn tqdm

jupyter notebook MNIST.ipynb

✅ Model checkpoints saved in the models/ directory. 🖼️ Confusion matrix saved in the images/ folder.

The notebook includes the following architectures:

• 🔹 Linear Model using nn.Linear
• 🔹 Convolutional Neural Network (CNN) using nn.Conv2d

• MNIST digit classification using PyTorch
• Custom training and validation loops
• Live progress bar via tqdm
• Evaluation with F1 Score and Accuracy
• Confusion matrix visualization
• GPU acceleration support

Linear Model        CNN Model