Imported from chrishuffman5/powershell (.codex/skills/aws-cli/sagemaker/SKILL.md). Install upstream with npx skills add chrishuffman5/powershell --skill sagemaker. Copyright stays with the author.

AWS CLI v2 — Amazon SageMaker

Overview

Complete reference for aws sagemaker and aws sagemaker-runtime subcommands in AWS CLI v2. Covers training jobs, model creation, endpoint deployment, processing jobs, hyperparameter tuning, AutoML, pipelines, experiments, Feature Store, model registry, model cards, monitoring, inference experiments, notebook instances, HyperPod clusters, Studio domains, MLflow integration, optimization jobs, labeling, workforce management, custom images, SageMaker Hubs, lineage tracking, edge deployment, partner apps, studio lifecycle configs, batch transform, algorithms, projects, code repositories, search, and real-time/async inference.

Quick Reference — Common Workflows

Create a training job

aws sagemaker create-training-job \
  --training-job-name my-training \
  --role-arn arn:aws:iam::123456789012:role/SageMakerRole \
  --algorithm-specification TrainingImage=382416733822.dkr.ecr.us-east-1.amazonaws.com/xgboost:latest,TrainingInputMode=File \
  --input-data-config '[{"ChannelName":"train","DataSource":{"S3DataSource":{"S3Uri":"s3://bucket/train","S3DataType":"S3Prefix"}}}]' \
  --output-data-config S3OutputPath=s3://bucket/output \
  --resource-config InstanceType=ml.m5.xlarge,InstanceCount=1,VolumeSizeInGB=30 \
  --stopping-condition MaxRuntimeInSeconds=86400

Create a model and deploy to an endpoint

aws sagemaker create-model \
  --model-name my-model \
  --primary-container Image=382416733822.dkr.ecr.us-east-1.amazonaws.com/xgboost:latest,ModelDataUrl=s3://bucket/output/model.tar.gz \
  --execution-role-arn arn:aws:iam::123456789012:role/SageMakerRole

aws sagemaker create-endpoint-config \
  --endpoint-config-name my-config \
  --production-variants '[{"VariantName":"primary","ModelName":"my-model","InstanceType":"ml.m5.xlarge","InitialInstanceCount":1}]'

aws sagemaker create-endpoint \
  --endpoint-name my-endpoint \
  --endpoint-config-name my-config

aws sagemaker wait endpoint-in-service --endpoint-name my-endpoint

Invoke an endpoint

aws sagemaker-runtime invoke-endpoint \
  --endpoint-name my-endpoint \
  --content-type text/csv \
  --body '1.0,2.0,3.0' \
  output.json

Create and run a pipeline

aws sagemaker create-pipeline \
  --pipeline-name my-pipeline \
  --pipeline-definition file://pipeline.json \
  --role-arn arn:aws:iam::123456789012:role/SageMakerRole

aws sagemaker start-pipeline-execution \
  --pipeline-name my-pipeline

Run a processing job

aws sagemaker create-processing-job \
  --processing-job-name my-processing \
  --role-arn arn:aws:iam::123456789012:role/SageMakerRole \
  --processing-resources '{"ClusterConfig":{"InstanceCount":1,"InstanceType":"ml.m5.xlarge","VolumeSizeInGB":30}}' \
  --app-specification ImageUri=382416733822.dkr.ecr.us-east-1.amazonaws.com/sagemaker-scikit-learn:latest

Create a hyperparameter tuning job

aws sagemaker create-hyper-parameter-tuning-job \
  --hyper-parameter-tuning-job-name my-tuning \
  --hyper-parameter-tuning-job-config '{"Strategy":"Bayesian","HyperParameterTuningJobObjective":{"Type":"Maximize","MetricName":"validation:auc"},"ResourceLimits":{"MaxNumberOfTrainingJobs":20,"MaxParallelTrainingJobs":3}}' \
  --training-job-definition file://training-def.json

Create a Feature Store feature group

aws sagemaker create-feature-group \
  --feature-group-name my-features \
  --record-identifier-feature-name customer_id \
  --event-time-feature-name event_time \
  --feature-definitions '[{"FeatureName":"customer_id","FeatureType":"String"},{"FeatureName":"event_time","FeatureType":"String"},{"FeatureName":"age","FeatureType":"Integral"}]' \
  --online-store-config '{"EnableOnlineStore":true}' \
  --role-arn arn:aws:iam::123456789012:role/SageMakerRole

Register a model in Model Registry

aws sagemaker create-model-package \
  --model-package-group-name my-model-group \
  --inference-specification '{"Containers":[{"Image":"382416733822.dkr.ecr.us-east-1.amazonaws.com/xgboost:latest","ModelDataUrl":"s3://bucket/model.tar.gz"}],"SupportedTransformInstanceTypes":["ml.m5.xlarge"],"SupportedRealtimeInferenceInstanceTypes":["ml.m5.xlarge"],"SupportedContentTypes":["text/csv"],"SupportedResponseMIMETypes":["text/csv"]}' \
  --model-approval-status Approved

Command Reference

See index.md for the quick reference table and global options.