jerqi commented on code in PR #54: URL: https://github.com/apache/gravitino-playground/pull/54#discussion_r1673793067
########## README.md: ########## @@ -224,6 +224,35 @@ union select * from catalog_iceberg.sales.customers; ``` +### Using Gravitino with LlamaIndex + +Gravitino playground also provides a simple RAG demo with LlamaIndex. This demo will show you the +ability of using Gravitino to manage both tabular and non-tabular dataset, connecting to +LlamaIndex as a unified data source, then use LlamaIndex and LLM to query both tabular and +non-tabular data with one natural language query. + +The demo is located in the `jupyter` folder, you can open the `gravitino_llama_index_demo.ipynb` +demo via Jupyter Notebook by [http://localhost:8888](http://localhost:8888). + +The scenario of this demo is that basic structured city statistics data is stored in MySQL, and +detailed city introductions are stored in PDF files. The user wants to know the answers to the +cities both in the structured data and the PDF files. + +In this demo, we will use Gravitino to manage the MySQL table using relational catalog, pdf Review Comment: `we will` -> `you can` -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
