MNIST Digit Classification with PyTorch

This project implements a simple fully connected neural network (MLP) to classify handwritten digits from the MNIST dataset using PyTorch.

The network consists of four fully connected layers with ReLU activations and is trained using the Adam optimizer.

---

Features

• Fully connected neural network (MLP) for MNIST
• Uses PyTorch and torchvision
• Visualizes sample predictions
• Outputs test accuracy
• Easy to modify for experimenting with architectures

---

Requirements

• Python 3.6+
• PyTorch
• Torchvision
• Matplotlib

Install dependencies using pip:

pip install torch torchvision matplotlib