Back to Main Page
Neural Network from Scratch
Technologies
C++Eigen
Overview
This project implements a neural network from scratch using only C++ and the Eigen library for linear algebra.
Key Features
- Implemented classes for a fully connected feedforward neural network
- Used the Eigen library for efficient linear algebra operations
- Implemented forward and backward propagation
- Currently working on adding support for convolutional layers
Outcomes
- Successfully implemented a fully connected feedforward neural network trained on the MNIST Digits dataset with 96% accuracy