Subclassing Feed-Forward Neural Network (FFN) for MNIST & Fashion MNIST classification.

Overview

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

Model Architecture

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

Custom Training

Results

After training for 10 epochs, the model achieves an accuracy of 98.25% on MNIST and 88.93% on Fashion MNIST test data.