Class Central is learner-supported. When you buy through links on our site, we may earn an affiliate commission.

Automating ML Workflows with SageMaker Pipelines

Go to class Write review

Overview

Supercharge your ML projects by building automated pipelines with SageMaker Pipelines. Connect data processing, training, evaluation, and deployment into robust, repeatable workflows. You’ll learn to monitor pipeline runs, add evaluation steps, and automate model registration and deployment—making your machine learning solutions faster, smarter, and easier to maintain.

Syllabus

Unit 1: Building Your First SageMaker Pipeline

Preparing Data for Pipeline Automation
Setting Up the Data Processor
Building Your First Processing Step
Completing SageMaker Processing Script Paths
Creating and Executing Your First Pipeline

Unit 2: Monitoring Pipeline Executions

Monitoring Your First Pipeline
Extracting Pipeline Execution Details
Calculating Pipeline Execution Duration
Monitoring Individual Pipeline Step Status
Comprehensive Step Timing and Error Handling

Unit 3: Integrating Model Training Steps

Configuring the SKLearn Estimator
Connecting Training to Data Processing
Updating Pipeline Output References
Monitoring Training Pipeline Execution

Unit 4: Evaluating Models in Pipelines

Configure Evaluation Property Files
Build Model Evaluation Pipeline Step
Connecting Pipeline Paths for Model Evaluation
Integrating the Evaluation Step to ML Pipeline
Monitor Your Complete Evaluation Pipeline

Unit 5: Conditionally Registering Models for Deployment

Creating Deployment Ready Model Packages
Registering Models in SageMaker Registry
Implementing Intelligent Model Quality Gates
Querying the Model Registry for Deployment
Deploying Approved Models to Production Endpoints

Reviews

Start your review of Automating ML Workflows with SageMaker Pipelines