Langchain callback fix for pyfunc serving #12023
Conversation
Signed-off-by: Serena Ruan <serena.rxy@gmail.com>
Signed-off-by: Serena Ruan <serena.rxy@gmail.com>
Signed-off-by: Serena Ruan <serena.rxy@gmail.com>
|
Documentation preview for 1393ce4 will be available when this CircleCI job More info
|
| @@ -641,7 +641,7 @@ def predict( | |||
| if is_in_databricks_model_serving_environment() and MLFLOW_ENABLE_TRACE_IN_SERVING.get(): | |||
| from mlflow.langchain.langchain_tracer import MlflowLangchainTracer | |||
|
|
|||
| callbacks = [MlflowLangchainTracer()] | |||
| callbacks = [MlflowLangchainTracer(prediction_context=get_prediction_context())] | |||
There was a problem hiding this comment.
This change is necessary for propagating request_id into processor in serving. But we need to set prediction_context in serving
Signed-off-by: Serena Ruan <serena.rxy@gmail.com>
|
This PR is not a blocker for rag callback migration now, but it should be helpful for mlflow serving tracing. |
| # temp workaround for rag model in model serving passing convert_chat_responses | ||
| # to predict method, we shouldn't validate its schema | ||
| if self.loader_module == "mlflow.langchain": | ||
| convert_chat_responses = params.pop("convert_chat_responses", None) if params else None | ||
| else: | ||
| convert_chat_responses = None | ||
| params = _validate_params(params, self.metadata) | ||
| if convert_chat_responses is not None: | ||
| params["convert_chat_responses"] = convert_chat_responses |
There was a problem hiding this comment.
Instead, I think we should start allowing parameters that aren't in the signature to be passed. Can we make that change instead? I'm concerned about adding special case logic in pyfunc predict for specific flavors. If this misses the 2.13.0 release, that's okay.
There was a problem hiding this comment.
Ok, I think that makes sense for allowing more params skipping params validation, it's just currently convert_chat_responses must be set to true so it returns a dictionary as expected. We can revisit this PR later for general mlflow pyfunc tracing support.
There was a problem hiding this comment.
Instead, I think we should start allowing parameters that aren't in the signature to be passed.
Then we can also add the callback as the parameters ?
There was a problem hiding this comment.
what about allowing _validate_prediction_input to be a overridable method in model_impl ? instead of adding a patch here.
🛠 DevTools 🛠
Install mlflow from this PR
Checkout with GitHub CLI
Related Issues/PRs
#xxxWhat changes are proposed in this pull request?
Fix langchain tracer to use prediction_context, and support passing convert_chat_responses as params.
How is this PR tested?
Does this PR require documentation update?
Release Notes
Is this a user-facing change?
What component(s), interfaces, languages, and integrations does this PR affect?
Components
area/artifacts: Artifact stores and artifact loggingarea/build: Build and test infrastructure for MLflowarea/deployments: MLflow Deployments client APIs, server, and third-party Deployments integrationsarea/docs: MLflow documentation pagesarea/examples: Example codearea/model-registry: Model Registry service, APIs, and the fluent client calls for Model Registryarea/models: MLmodel format, model serialization/deserialization, flavorsarea/recipes: Recipes, Recipe APIs, Recipe configs, Recipe Templatesarea/projects: MLproject format, project running backendsarea/scoring: MLflow Model server, model deployment tools, Spark UDFsarea/server-infra: MLflow Tracking server backendarea/tracking: Tracking Service, tracking client APIs, autologgingInterface
area/uiux: Front-end, user experience, plotting, JavaScript, JavaScript dev serverarea/docker: Docker use across MLflow's components, such as MLflow Projects and MLflow Modelsarea/sqlalchemy: Use of SQLAlchemy in the Tracking Service or Model Registryarea/windows: Windows supportLanguage
language/r: R APIs and clientslanguage/java: Java APIs and clientslanguage/new: Proposals for new client languagesIntegrations
integrations/azure: Azure and Azure ML integrationsintegrations/sagemaker: SageMaker integrationsintegrations/databricks: Databricks integrationsHow should the PR be classified in the release notes? Choose one:
rn/none- No description will be included. The PR will be mentioned only by the PR number in the "Small Bugfixes and Documentation Updates" sectionrn/breaking-change- The PR will be mentioned in the "Breaking Changes" sectionrn/feature- A new user-facing feature worth mentioning in the release notesrn/bug-fix- A user-facing bug fix worth mentioning in the release notesrn/documentation- A user-facing documentation change worth mentioning in the release notesShould this PR be included in the next patch release?
Yesshould be selected for bug fixes, documentation updates, and other small changes.Noshould be selected for new features and larger changes. If you're unsure about the release classification of this PR, leave this unchecked to let the maintainers decide.What is a minor/patch release?
Bug fixes, doc updates and new features usually go into minor releases.
Bug fixes and doc updates usually go into patch releases.