Genai: Integrate JSON Mode for Gemini 1.5 (flash) #625
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
vishah02
changed the title
|
@vishah02 thanks a lot for your contribution! could you fix linting errors and add an integration test, please? |
|
@lkuligin Hey, sorry it took a while to get to it. We added tests and fixed the linting errrors now! Let me know if there's anything else I can do! |
|
@vishah02 you can check linter locally if you'd like to: |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
PR Description
Updated ChatGoogleGenerativeAI to allow for outputs to be formatted into JSON mode. Added an optional parameter response_mime_type into ChatGoogleGenerativeAI that formats outputs into JSON mode when set to “application/json”, along with the ability to (optionally) provide schemas for the response to follow. Added support for enums in response_mime_type and response_schema, which ensures that the API’s response is from on of the values in the enum provided in response_schema
Relevant issues
#24184
Type
🆕 New Feature
Changes(optional)
All changes were made to libs\genai\langchain_google_genai\chat_models.py
Changes to validate_environment():
If response_mime_type and response_schema exist, ensure they are valid, else return an error
Valid response_mime_types include "text/plain", "application/json", "text/x.enum"
Response_schemas is only a valid field when used with response_mime_types "application/json" or "text/x.enum"
Changes to _prepare_params():
Add response_mime_type and response_schema to gen_config
If a response schema is provided, normalise/convert the schema to a protos_schema for the google API request
Testing(optional)
Manual Testing with Inputs
Passed all langchain-google-genai-us integration tests.
Note(optional)
Dependencies: imported generation_types from google.generativeai.types
Documentation: Added docstrings for JSON Output and Schema Support in class ChatGoogleGenerativeAI and for parameters response_mime_type and response_schema.