Image Classification with a Neural Network Built from Scratch using NumPy

Overview

This project demonstrates how to build and train a basic feedforward neural network using only NumPy. The model is designed to classify different fruits, such as apples, mangos, rasberries and lemons, based on their images from the Fruits-360 dataset.

Model Architecture

The neural network consists of:

• Input Layer: Takes the flattened image as input.
• Hidden Layers: Includes fully connected layers with ReLU activation.
• Output Layer: Uses softmax for multi-class classification.