https://github.com/cllspy/ai_from_scratch
learning AI from scratch by doing
https://github.com/cllspy/ai_from_scratch
Last synced: 5 months ago
JSON representation
learning AI from scratch by doing
- Host: GitHub
- URL: https://github.com/cllspy/ai_from_scratch
- Owner: CllsPy
- Created: 2025-09-11T14:21:37.000Z (7 months ago)
- Default Branch: main
- Last Pushed: 2025-09-19T01:57:51.000Z (6 months ago)
- Last Synced: 2025-09-19T03:42:03.720Z (6 months ago)
- Language: Jupyter Notebook
- Size: 56.6 KB
- Stars: 0
- Watchers: 0
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# AI From Scratch
A project dedicated to building artificial intelligence algorithms and models entirely from scratch. This repository aims to explore and implement fundamental AI concepts without relying heavily on pre-built libraries or frameworks.
## Overview
The purpose of this project is to understand the inner workings of AI by implementing algorithms manually. This includes, but is not limited to:
* Neural Networks
* Machine Learning algorithms (supervised and unsupervised)
* Reinforcement Learning
* Optimization techniques
* Data preprocessing and evaluation metrics
By building these systems from scratch, we aim to gain a deeper understanding of how AI works at its core.
## Features
* Implementation of core AI algorithms using Python (or preferred language)
* Step-by-step explanations of each algorithm
* Simple, modular code to encourage experimentation and learning
* Focus on clarity and understanding over performance
## Roadmap
* [ ] Linear regression from scratch
* [ ] Logistic regression
* [ ] Feedforward neural networks
* [ ] Backpropagation
* [ ] Convolutional neural networks
* [ ] Reinforcement learning basics
* [ ] Optimization algorithms (SGD, Adam, etc.)
* [X] Attention Mechanism
* [X] GPT
* [X] Ensemble (Bagging)