TinyNet

TinyNet is a straightforward implementation of a Feedforward Neural Network (FFNN) built from scratch in Python. Inspired by @karpathy's micrograd, this project aims to provide an educational resource for understanding the foundational components of neural networks without relying on external libraries.

Features

• Pure Python Implementation: Constructed without dependencies to offer clear insights into the mechanics of neural networks.
• Educational Focus: Designed to help students and enthusiasts grasp the inner workings of FFNNs through hands-on experience.
• Modular Design: Organized code structure for ease of understanding and extension.

Installation

Clone the repository to your local machine:

git clone https://github.com/nMaax/TinyNet.git
cd TinyNet

Usage

The primary code and examples are contained within the main.ipynb Jupyter Notebook. To explore and run the code:

1. Install Jupyter Notebook: If you don't have it installed, you can add it using pip:

   pip install notebook

2. Launch Jupyter Notebook:

   jupyter notebook

3. Open main.ipynb: In the Jupyter interface, navigate to the TinyNet directory and open main.ipynb.

4. Run the Notebook: Execute the cells sequentially to build and train the neural network.

Project Structure

• NN/: Contains the core neural network implementation.
• main.ipynb: Jupyter Notebook demonstrating the usage of the neural network.
• LICENSE: Project license information.
• .gitignore: Specifies files to ignore in the repository.

---

Note: This project is currently under development. Features and implementations are subject to change.