Subclassing Convolutional Neural Network (CNN) for MNIST & Fashion MNIST classification.

Overview

This repository contains a TensorFlow/Keras implementation of a Convolutional Neural Network (CNN) trained to classify MNIST & Fashion-MNIST dataset.

Model Architecture

The CNN model is defined using the subclassing method in TensorFlow/Keras. Here's a brief overview of the architecture:

Training

Both model is trained using the Adam optimizer with sparse categorical cross-entropy loss.

Results

After training for 10 epochs, the model achieves an accuracy of 99.45% on MNIST & 92.12% on Fashion MNIST.