Fashion MNIST Classification

This project contains deep learning models built using PyTorch to classify images from the Fashion MNIST dataset. The goal is to accurately recognize different categories of clothing items using both simple and custom Convolutional Neural Network (CNN) architectures.

📂 Dataset

The Fashion MNIST dataset is a drop-in replacement for the classic MNIST dataset and contains 60,000 training and 10,000 test grayscale images of 10 clothing categories.

🧠 Models Implemented

• Simple Fully Connected Neural Network (FCNN)
• Custom CNN
• Tiny VGG16-inspired CNN (https://poloclub.github.io/cnn-explainer/)

📊 Features

• Training and evaluation on Fashion MNIST
• Model comparison and performance metrics
• Plotting of loss and accuracy over epochs

🛠️ Technologies Used

• Python
• PyTorch
• Matplotlib
• NumPy
• Jupyter Notebook