mypy-boto3-sagemaker-metrics type annotations

Warnings

• breaking As of `mypy-boto3-builder` version 8.12.0 (which generated `mypy-boto3-sagemaker-metrics` 1.42.3), Python 3.8 support has been explicitly removed. Projects requiring type stubs for this service must use Python 3.9 or newer.
  Fix: Upgrade your project's Python version to 3.9 or later, or pin an older version of `mypy-boto3-sagemaker-metrics` that supported Python 3.8 if absolute compatibility is needed.
• breaking In `mypy-boto3-builder` 8.9.0, there were breaking changes to TypeDef naming conventions. Specifically, suffixes like `RequestRequestTypeDef` were shortened to `RequestTypeDef`, and `Extra` postfixes were moved. While this specific package's service might not be directly affected, other `mypy-boto3` packages used in conjunction could break.
  Fix: Review your code for direct imports of TypeDefs from `mypy-boto3` packages and adjust names according to the new conventions, often found in the `type_defs` submodule.
• gotcha This package provides *only* type annotations for `boto3`. It does not include the `boto3` library itself. You must install `boto3` separately for your code to run at runtime.
  Fix: Ensure `boto3` is included in your project's dependencies: `pip install boto3`.
• gotcha The `mypy-boto3` libraries migrated to PEP 561 packages. While this generally improves compatibility with type checkers, older `mypy` configurations or non-standard `pip` usage might require explicit configuration or ensuring packages are installed in the environment where `mypy` runs.
  Fix: Verify your `mypy` configuration (e.g., `pyproject.toml`) is up-to-date and that `mypy-boto3-sagemaker-metrics` is installed in the Python environment used by your type checker. For best results, use `uv` or `pip` for installation.

Install

• pip install mypy-boto3-sagemaker-metrics Install the type stubs
• pip install boto3 Install the runtime library (required)

Imports

• SageMakerMetricsClient

  from mypy_boto3_sagemaker_metrics.client import SageMakerMetricsClient
  import boto3
  
  client: SageMakerMetricsClient = boto3.client("sagemaker-metrics")

  The type stub provides the type hint for the client returned by `boto3.client`.
• BatchPutMetricsResponseTypeDef

  from mypy_boto3_sagemaker_metrics.type_defs import BatchPutMetricsResponseTypeDef

  TypedDicts are provided for request/response bodies for enhanced type checking.

Quickstart

This quickstart demonstrates how to use the type-hinted SageMakerMetrics client to put batch metrics. It shows the correct way to instantiate a client with explicit type annotation and then call a service method, benefiting from IDE auto-completion and static type checks. Replace `my-trial-component-name` with an actual SageMaker Trial Component name for execution.

import boto3
from datetime import datetime
from mypy_boto3_sagemaker_metrics.client import SageMakerMetricsClient


def put_example_metrics(trial_component_name: str):
    client: SageMakerMetricsClient = boto3.client("sagemaker-metrics")

    response = client.batch_put_metrics(
        TrialComponentName=trial_component_name,
        MetricData=[
            {
                'MetricName': 'accuracy',
                'Timestamp': datetime.now(),
                'Step': 0,
                'Value': 0.95
            },
            {
                'MetricName': 'loss',
                'Timestamp': datetime.now(),
                'Step': 0,
                'Value': 0.05
            },
        ]
    )

    print(f"Successfully put metrics: {response}")

# Example usage (requires an existing SageMaker Trial Component)
# try:
#     put_example_metrics("my-trial-component-name")
# except Exception as e:
#     print(f"Error putting metrics: {e}")

# Note: Creating a Trial Component is outside the scope of this quickstart.
# You would typically get this name from a SageMaker training job or experiment.