RAG Tutorial Notebooks

Note

content/it/rag/notebooks

These notebooks guide you through implementing Retrieval Augmented Generation (RAG) pipelines.

1. 0_download_data.ipynb: Download and prepare data for RAG pipelines

2. 1_chunking_indexing_data.ipynb: Data chunking, embedding, and vector index creation

3. 2_creating_a_chatbot.ipynb: Build a complete RAG-powered chatbot

Notebooks

Download Data

• General
• Imports
• Env config
• Scraping
• Download data

Chunking and indexing data

• General
• Imports
• Env config
• Why do we need RAG systems?
• Classical information retrieval approaches
• Semantic search
• Finding the right chunk size
• Chunk and index all documents
• Evaluating k-size in the retrieval step
• The final retriever
• Improving the system
• A combined approach
• Contextual retrieval
• A few considerations about contextual retrieval
• Final remarks

Creating a chatbot

• General
• Imports
• Env config
• The collections
• Workflow
• Agents and agentic RAG
• Considerations
• Langchain
• Next steps