Neural Networks from Scratch

This repository contains Jupyter Notebooks with hands-on implementations to help you build neural networks from scratch. The goal is to explore and understand the core principles of neural networks by implementing and experimenting with basic architectures, activation functions, and learning algorithms.

Structure

• basic-neural-networks: Contains experiments focused on the fundamentals of neural networks, including individual neurons, perceptrons, and simple problem applications.
• activation-functions: Explores various activation functions and their implementations in neural networks, such as Sigmoid, Tanh, ReLU, and Softmax.

Additional Folders: More advanced topics and architectures will be added as the learning journey progresses.

Requirements

• Python 3.x
• NumPy
• Matplotlib
• Jupyter Notebook or Google Colab

To install the required dependencies:

pip install numpy matplotlib