Interface ModelServiceGrpc.AsyncService (2.16.0)

public static interface ModelServiceGrpc.AsyncService

Service for performing CRUD operations on models. Recommendation models contain all the metadata necessary to generate a set of models for the Predict() API. A model is queried indirectly via a ServingConfig, which associates a model with a given Placement (e.g. Frequently Bought Together on Home Page). This service allows you to do the following:

  • Initiate training of a model.
  • Pause training of an existing model.
  • List all the available models along with their metadata.
  • Control their tuning schedule.

Methods

createModel(CreateModelRequest request, StreamObserver<Operation> responseObserver)

public default void createModel(CreateModelRequest request, StreamObserver<Operation> responseObserver)

Creates a new model.

Parameters
NameDescription
requestCreateModelRequest
responseObserverio.grpc.stub.StreamObserver<Operation>

deleteModel(DeleteModelRequest request, StreamObserver<Empty> responseObserver)

public default void deleteModel(DeleteModelRequest request, StreamObserver<Empty> responseObserver)

Deletes an existing model.

Parameters
NameDescription
requestDeleteModelRequest
responseObserverio.grpc.stub.StreamObserver<Empty>

listModels(ListModelsRequest request, StreamObserver<ListModelsResponse> responseObserver)

public default void listModels(ListModelsRequest request, StreamObserver<ListModelsResponse> responseObserver)

Lists all the models linked to this event store.

Parameters
NameDescription
requestListModelsRequest
responseObserverio.grpc.stub.StreamObserver<ListModelsResponse>

pauseModel(PauseModelRequest request, StreamObserver<Model> responseObserver)

public default void pauseModel(PauseModelRequest request, StreamObserver<Model> responseObserver)

Pauses the training of an existing model.

Parameters
NameDescription
requestPauseModelRequest
responseObserverio.grpc.stub.StreamObserver<Model>

resumeModel(ResumeModelRequest request, StreamObserver<Model> responseObserver)

public default void resumeModel(ResumeModelRequest request, StreamObserver<Model> responseObserver)

Resumes the training of an existing model.

Parameters
NameDescription
requestResumeModelRequest
responseObserverio.grpc.stub.StreamObserver<Model>

tuneModel(TuneModelRequest request, StreamObserver<Operation> responseObserver)

public default void tuneModel(TuneModelRequest request, StreamObserver<Operation> responseObserver)

Tunes an existing model.

Parameters
NameDescription
requestTuneModelRequest
responseObserverio.grpc.stub.StreamObserver<Operation>

updateModel(UpdateModelRequest request, StreamObserver<Model> responseObserver)

public default void updateModel(UpdateModelRequest request, StreamObserver<Model> responseObserver)

Update of model metadata. Only fields that currently can be updated are: filtering_option and periodic_tuning_state. If other values are provided, this API method ignores them.

Parameters
NameDescription
requestUpdateModelRequest
responseObserverio.grpc.stub.StreamObserver<Model>