Skip to content

Enhance documentation for stateful conversations #38247

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
damccorm wants to merge 1 commit into
base: master
Choose a base branch
from
+6 −1
Open

Enhance documentation for stateful conversations #38247

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
7 changes: 6 additions & 1 deletion sdks/python/apache_beam/ml/inference/agent_development_kit.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -96,7 +96,12 @@ class ADKAgentModelHandler(ModelHandler[str | genai_Content,
batch. By default every invocation uses a fresh, isolated session (stateless).
Stateful multi-turn conversations can be achieved by passing a ``session_id``
key inside ``inference_args``; elements sharing the same ``session_id`` will
continue the same conversation history.
continue the same conversation history. When using stateful conversations,
it is recommended to use a custom session_service_factory to provide a session
service implementation which can be managed across multiple workers (e.g.
:class:`~google.adk.sessions.DatabaseSessionService`). The default
:class:`~google.adk.sessions.InMemorySessionService` will not correctly track
the same session across multiple workers.

Args:
agent: A pre-constructed :class:`~google.adk.agents.Agent` instance, or a
Expand Down
Loading