Skip to content

Latest commit

 

History

History
18 lines (8 loc) · 1.24 KB

README.md

File metadata and controls

18 lines (8 loc) · 1.24 KB

Mailing list : test License: CC BY-NC 4.0

cgpt

LangChain OpenAI based QnA on the Table.

This project demonstrates the use of ChatGPT models to query tabular data and obtain accurate answers.

Instead of directly feeding the table content to the model, we utilize the langchain feature to prompt the model to generate a relevant query for the database.

The query results are then summarized by the model to provide the answer.

This approach improves efficiency and accuracy in retrieving information from tabular data. It leverages established techniques in natural language processing and database management. The project incorporates principles of information retrieval and summarization, ensuring the answers are concise and comprehensive.

  • You would require OpenAI API key to run this sample. Key can be set as environment variable in the OS or can be added to .env file and read.