winterbaume-sagemakermetrics

SageMaker Metrics service implementation for winterbaume.

Coverage

Metric	Value
Service	SageMaker Metrics
AWS model	sagemaker-metrics
Protocol	restJson1
winterbaume coverage	2/2 operations (100.0%)
stubs (routed, returns empty/default)	0/2 operations (0.0%)
moto coverage	1/2 operations (50.0%)
floci coverage	0/2 operations (0.0%)
kumo coverage	0/2 operations (0.0%)
Coverage report date	2026-05-13

Server-mode usage

Start winterbaume-server and point the AWS CLI at it:

cargo run -p winterbaume-server -- --host 127.0.0.1 --port 5555

export AWS_ENDPOINT_URL=http://localhost:5555
aws sagemaker-metrics batch-get-metrics --metric-queries MetricName=loss,ResourceArn=arn:aws:sagemaker:us-east-1:123456789012:experiment-trial/my-trial,MetricStat=Max,Period=OneMinute,XAxisType=TIMESTAMP

Example

use aws_sdk_sagemakermetrics::config::BehaviorVersion;
use winterbaume_core::MockAws;
use winterbaume_sagemakermetrics::SageMakerMetricsService;

#[tokio::main]
async fn main() {
    let mock = MockAws::builder()
        .with_service(SageMakerMetricsService::new())
        .build();

    let config = aws_config::defaults(BehaviorVersion::latest())
        .http_client(mock.http_client())
        .credentials_provider(mock.credentials_provider())
        .region(aws_sdk_sagemakermetrics::config::Region::new("us-east-1"))
        .load()
        .await;

    let client = aws_sdk_sagemakermetrics::Client::new(&config);

    // SageMaker Metrics requires training job or experiment run context.
    // This example demonstrates client setup for the SageMakerMetrics service.
    println!("SageMakerMetrics client ready. Use batch_put_metrics() to record training metrics.");
    let _client = client;
}

Implemented APIs (2)

• BatchGetMetrics
• BatchPutMetrics