Applications of Foundation Models
SageMaker Platform
Amazon SageMaker is the managed ML platform for building, training, tuning, deploying, and monitoring models. For AIF-C01, focus on which SageMaker feature maps to no-code ML, labeling, bias/explainability, monitoring, feature management, and pipelines.
Concepts
- Amazon SageMaker is a managed ML platform for building, training, tuning, deploying, and monitoring models.
- SageMaker JumpStart provides prebuilt models and solution templates.
- SageMaker Canvas supports no-code ML and generative AI workflows.
- SageMaker Autopilot automates model building from tabular data.
- SageMaker Ground Truth supports human and automated data labeling.
- SageMaker Clarify detects bias and explains model predictions.
- SageMaker Model Monitor detects data quality, model quality, bias, and explainability drift.
- SageMaker Data Wrangler prepares data; Feature Store manages reusable ML features; Pipelines orchestrate ML workflows; Debugger inspects training; Automatic Model Tuning optimizes hyperparameters.
Exam tips
- SageMaker Canvas is no-code ML; Autopilot automates tabular model building.
- Ground Truth supports data labeling; Clarify supports bias detection and explainability.
- Model Monitor tracks deployed model quality, data quality, bias, and explainability drift.