Gradient Maker

Gradient Maker is a lightweight implementation demonstrating the core principles of gradient descent. This repository is designed to help users understand how gradient descent works under the hood without relying on heavy frameworks.

---

Overview

Gradient descent is the backbone of machine learning training. This project uses a simple Python implementation to show the concepts behind gradient descent by leveraging a graph-based approach to store and propagate gradients through computational nodes.

---

Features

• Minimal Dependencies: Requires only basic Python knowledge and high school level calculus.

---

Repository Structure

• nn.py: Contains an example of a Multi-Layer Perceptron (MLP) that trains on a simple dataset.
• gradient_maker.py: Defines classes and functions to implement gradient descent using a computational graph.
• README.md: Provides an overview of the project, its purpose, and usage instructions.
• .gitignore: Lists files and directories to be ignored by Git.

---

How to Use

Running the MLP Example

1. Clone the Repository:

   git clone https://github.com/KrishnaAgarwal1308/gradinet_maker.git
   cd gradinet_maker
   python nn.py

Customizing Gradient Descent

Feel free to modify the code in gradient_maker.py to experiment with different functions, gradient computations, or integrate more complex models.

Prerequisites

Python 3.x: Ensure you have Python installed on your machine.

Basic Python Knowledge: Understanding object-oriented programming in Python will help.

High School Level Calculus: Familiarity with basic calculus concepts such as derivatives is recommended.