This repository provides an 8-week curriculum for learning AI and Machine Learning, targeting coders and developers. It offers a structured path through core ML concepts, computer vision, natural language processing, and tabular data analysis, culminating in AI ethics and prototype deployment, with the goal of building practical AI applications.

How It Works

The curriculum is structured into weekly modules, each focusing on specific AI/ML topics. It leverages popular libraries like PyTorch, FastAI, and Hugging Face, providing hands-on experience through code notebooks. The approach includes building foundational concepts from scratch (e.g., Stochastic Gradient Descent) and applying transfer learning for complex tasks like image classification and object detection. Key areas like NLP are covered using Transformer models and LLMs, while tabular data analysis includes regression and classification.

Quick Start & Requirements

• Installation: Primarily involves running provided Jupyter notebooks.
• Prerequisites: Python, PyTorch, FastAI, Hugging Face libraries (transformers, datasets, tokenizers), torchvision, Streamlit. Access to YouTube for video lectures and Google Slides/PDFs for slides is recommended.
• Resources: Requires a development environment capable of running Python and ML libraries. GPU acceleration is beneficial for vision and LLM tasks.
• Links:
  • Videos: Linked per lesson on YouTube.
  • Notebooks: Linked per lesson.
  • Slides: Linked per lesson (Google Slides/PDF).
  • Fastai Course: course.fast.ai

Highlighted Details

• Covers foundational ML concepts to advanced topics like GANs, Transformers, and LLMs.
• Includes practical tracks for Vision, Text, and Tabular Data.
• Emphasizes AI ethics and responsible development.
• Guides users through deploying ML prototypes using Streamlit.
• Adapts and augments content from fastai's "Practical Deep Learning for Coders v4".

Maintenance & Community

• The project is maintained by ai-builders.
• Content is adapted from fastai, indicating a reliance on that project's ecosystem.

Licensing & Compatibility

• The curriculum content is adapted from fastai, which is licensed under GNU General Public License v3.0 (GPL-3.0). This license is copyleft and may have implications for commercial use or integration into closed-source projects.

Limitations & Caveats

The curriculum's reliance on GPL-3.0 licensed material may impose copyleft restrictions on derivative works or projects that incorporate its code or content, requiring careful consideration for commercial applications.