Grounding vertex AI Gemini

[jmugicagonz]

Description

Please include a summary of the change and which issue is fixed. Please also include relevant motivation and context. List any dependencies that are required for this change.

Fixes # (issue)

Type of Change

Please delete options that are not relevant.

• Bug fix (non-breaking change which fixes an issue)
• New feature (non-breaking change which adds functionality)
• Breaking change (fix or feature that would cause existing functionality to not work as expected)
• This change requires a documentation update

How Has This Been Tested?

Tested functionalities for chat, passing datastore as string parameter.

Please describe the tests that you ran to verify your changes. Provide instructions so we can reproduce. Please also list any relevant details for your test configuration

• Added new unit/integration tests
• Added new notebook (that tests end-to-end)
• I stared at the code and made sure it makes sense

Suggested Checklist:

• [X ] I have performed a self-review of my own code
• I have commented my code, particularly in hard-to-understand areas
• I have made corresponding changes to the documentation
• I have added Google Colab support for the newly added notebooks.
• [X ] My changes generate no new warnings
• [X ] I have added tests that prove my fix is effective or that my feature works
• New and existing unit tests pass locally with my changes
• I ran make format; make lint to appease the lint gods

[logan-markewich]

Interesting. Do you mind giving me a quick rundown of what this does to API calls? Does it force the LLM to act as an agent with this single retrieval tool?

Does vertexai support tools for any LLM? This would be really nice for agents for example, similar to openais tools api

[jmugicagonz]

Hello Logan,

Thanks for your answer. What this does is, in case you have a Verte datastore defined (see this link) you can ground your model in the data store. This is incredibly convenient for RAG applications, as creating a datastore is very easy.

And it doesn't force the LLM to act as an agent. You are just giving the LLM your datastore as a resource. You can tell in the prompt to give an answer strictly using the datastore, to reduce hallucinations.

[sandangel]

hmm, linting for 3.9 is failing so auto merge not happen yet. :(