Skip to content

A hands on lab project creating a retrieval agumented generation (RAG) app

Notifications You must be signed in to change notification settings

heinko/databricks-chatbot-llm

Repository files navigation

Databricks demo in Hands on workshop at Singapore

Processes

  • Process the .pdf file into Databricks vector search. parse the text and split it into chunks. Chunks will then be embedded as vectors into Vector Search Index.
  • Prepare your Chatbot with LangChain. Use Langchain and Llama-2 to create a chatbot that will retrieve the text from Databricks Vector Search and answer your .pdf specific questions.
  • Plug it to Gradio App. wrap the Chatbot into a Gradio app, so you can share it and have a cool interface to use it

Databricks demos components

  • Delta Live Table
  • Data Science
  • Unity Catalog
  • BI/DW/DBSQL

About

A hands on lab project creating a retrieval agumented generation (RAG) app

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages