sagemaker.serve.model_server.multi_model_server.prepare

sagemaker.serve.model_server.multi_model_server.prepare#

Shared resources for prepare step of model deployment

Functions

prepare_for_mms(model_path, shared_libs, ...)

Prepares for InferenceSpec using model_path, writes inference.py, and captures dependencies to generate secret_key.

prepare_mms_js_resources(model_path, js_id)

Prepare serving when a JumpStart model id is given
sagemaker.serve.model_server.multi_model_server.prepare.prepare_for_mms(model_path: str, shared_libs: List[str], dependencies: dict, session: Session, image_uri: str, inference_spec: InferenceSpec | None = None) → str[source]#

Prepares for InferenceSpec using model_path, writes inference.py, and captures dependencies to generate secret_key.

Args:to

model_path (str) : Argument shared_libs (List[]) : Argument dependencies (dict) : Argument session (Session) : Argument inference_spec (InferenceSpec, optional) : Argument

(default is None)

Returns:

secret_key

Return type:

( str )

sagemaker.serve.model_server.multi_model_server.prepare.prepare_mms_js_resources(model_path: str, js_id: str, shared_libs: List[str] | None = None, dependencies: str | None = None, model_data: str | None = None) → tuple[source]#

Prepare serving when a JumpStart model id is given

Parameters:

  • model_path (str) – Argument

  • js_id (str) – Argument

  • shared_libs (List[]) – Argument

  • dependencies (str) – Argument

  • model_data (str) – Argument

Return type:

( str )

Contents