Udacity Generative AI Nanodegree: Personal Notes

These are my personal notes taken while following the Udacity Generative AI Nanodegree.

The Nanodegree asssumes basic data analysis skills with data science python libraries and databases, and has 4 modules that build up on those skills; each module has its corresponding folder in this repository with its guide Markdown file:

1. Generative AI Fundamentals: 01_Fundamentals_GenAI.
   • Foundation Models
   • Fine-Tuning
2. Large Language Models (LLMs) & Text Generation: 02_LLMs.
   • Transformers and LLMs
   • Retrieval Augmented Generation (RAG) Chatbots
3. Computer Vision and Generative AI: 03_ComputerVision.
   • Generative Adversarial Networks (GANs)
   • Vision Transformers
   • Diffusion Models
4. Building Generative AI Solutions: 04_BuildingSolutions.
   • Vector Databases
   • LangChain and Agents

Additionally, it is necessary to submit and pass some projects to get the certification:

• Project 1: Apply Lightweight Fine-Tuning to a Foundation Model - TBD.
• Project 2: Build Your Own Custom Chatbot - TBD.
• Project 3: AI Photo Editing with Inpainting - TBD.
• Project 4: Personalized Real Estate Agent - TBD.

Finally, also check some of my personal guides on related tools:

• My personal notes on the O'Reilly book Generative Deep Learning, 2nd Edition, by David Foster
• My personal notes on the O'Reilly book Natural Language Processing with Transformers, by Lewis Tunstall, Leandro von Werra and Thomas Wolf (O'Reilly)
• HuggingFace Guide: mxagar/tool_guides/hugging_face
• LangChain Guide: mxagar/tool_guides/langchain
• LLM Tools: mxagar/tool_guides/llms
• NLP Guide: mxagar/nlp_guide
• Deep Learning Methods for CV and NLP: mxagar/computer_vision_udacity/CVND_Advanced_CV_and_DL.md
• Deep Learning Methods for NLP: mxagar/deep_learning_udacity/DLND_RNNs.md

Setup

A regular python environment with the usual data science packages should suffice (i.e., scikit-learn, pandas, matplotlib, etc.); any special/additional packages and their installation commands are introduced in the guides. A recipe to set up a conda environment with my current packages is the following:

# Create the necessary Python environment
# NOTE: specific folders might require their own environment
# and have their own requirements.txt
conda create --name ds pip python=3.10
conda activate ds
pip install -r requirements.txt

# When the repository is cloned, initialize and update the submodules 
git clone https://github.com/mxagar/generative_ai_udacity
git submodule update --init --recursive

Interesting Links

• microsoft/generative-ai-for-beginners
• ...

Credits

Many of the contents in this repository were created following the Udacity Generative AI Nanodegree.

Mikel Sagardia, 2024.
No guarantees.