This repository contains a collection of Jupyter notebooks designed to provide hands-on experience with various advanced AI techniques and tools. This is the source code used for the hands on exercises used for the AI Microclasses course held by the TUM-VentureLabs.
This workshop series covers a wide range of topics essential for understanding and implementing modern AI solutions. The notebooks are structured to guide participants through practical examples and detailed explanations. Links to collab notebooks can be found:
- https://colab.research.google.com/drive/1Ag_2AwwIRFrtygmlk_HIw-Pb1YyQLrcN?usp=sharing
- https://colab.research.google.com/drive/1uCMHYFJKtcLroslLDWaLFafc-ZgInwrJ?usp=sharing
- https://colab.research.google.com/drive/19jL1IgpmT2RNEqkJaHpOiqISAvXUE2Yd?usp=sharing
Below is a summary of the key topics covered:
Learn how to interact with various APIs programmatically. This section includes:
- Basics of making API requests
- Handling different types of API responses (JSON, Images, Text etc.)
- Error handling and debugging
Explore the fundamentals of web scraping to extract data from websites. Topics include:
- Using libraries like BeautifulSoup
- Navigating and parsing HTML documents
Understand how to use LangChain for language model manipulation. This includes:
- Wrapping and interfacing with different language models
- Customizing language model behavior
- Implementing advanced language model workflows
- Creating dynamic prompts with Prompt Templates
- Creating complex models using Chaining
- Structuring output into desired formats
- Splitting large texts into sizeable chunks
- Model Embedding and vectorization
- Retrieval-Augmented Generation
The notebooks in this repository are designed to provide practical, hands-on experience with the above topics. They are intended for participants of "AI Microclasses" workshops, and anyone aiming to learn. The goal is to equip you with the knowledge and skills needed to implement and utilize advanced AI techniques in real-world applications.
We welcome contributions to improve these materials. If you have suggestions, improvements, or additional examples, please submit a pull request or open an issue.