What you will Learn
Embeddings
- What are embeddings?
- How do you convert text into embedding vectors?
- Practical: Create embeddings of PDF file
Vector database (ChromaDB, Pinecone)
- Introduction to Vector DB
- Insert embeddings into Vector DB
- Perform CRUD operation on Vector DB
What is Retrieval-Augmented Generation (RAG)?
What distinguishes Semantic Search from Generative Search?
RAG utilizing Embeddings, Vector Database, Semantic search, and LLM Recommended strategies for RAG:
- Integration of semantic search and re-ranking
- Implementation of caching to decrease LLM costs and latency
- Continuous training and education
Implementing RAG through LangChain.
- Developers can easily construct LLM-powered apps using LangChain's framework
Implementing RAG through LangChain.
- A straightforward and adaptable data framework that links custom data sources to expansive language models
Model Evaluation
Project 1: Multiple PDF Bot by using RAG, LangChain
The goal is to develop a questionanswering bot capable of extracting relevant information from multiple PDF documents. Leveraging the RAG (Retrieval-Augmented Generation) model and LangChain for transcription, the bot will retrieve passages from the PDFs, generate answers to user questions, and provide accurate responses, streamlining information retrieval and enhancing user experience.
Project 2: YouTube Playlist QnA Bot by using RAG, LLaMA Index
The "YouTube Playlist QnA Bot" aims to address the challenge of efficiently extracting relevant information from YouTube playlists. Leveraging Generative Al, particularly the RAG (RetrievalAugmented Generation) model and LLaMA Index, the bot will retrieve and generate answers to user queries based on the content of YouTube playlists, enabling seamless interaction and retrieval of information from within the platform.
Project 3: Wikipedia QnA Bot
The Wikipedia QnA Bot aims to provide accurate and informative responses to user queries by leveraging generative Al technology. By analyzing vast amounts of Wikipedia articles, the bot generates contextually relevant answers to questions asked by users, facilitating easy access to information and enhancing the user experience.
What to Expect
Hands-on Training: Gain practical experience with industry-leading tools and
techniques.
Expert Guidance: Learn from seasoned professionals with years of experience
in the field of Al.
Collaborative Environment: Engage with like-minded individuals and foster
creative collaboration.
Real-World Applications: Discover how Generative Al is revolutionizing various
industries, from art and design to healthcare and beyond.
Who Should join
- Professionals seeking to enhance their skills and stay ahead in their careers.
- Students are eager to explore the intersection of technology and creativity.
- Anyone with a passion for innovation and a desire to make an impact with Al.
Instructors
- Our cohort will be led by experienced working Al professionals with expertise in Generative Al and deep learning
Certification
- Upon successful completion of the cohort, participants will receive a certificate of completion
Prerequisites
- Basic knowledge of Python programming language
- Familiarity with machine learning concepts is helpful but not required