Use Amazon SageMaker to build high-quality ML models faster
summary
Live demo shows SageMaker Canvas building tabular models (standard build takes about 1–3 hours) by iterating SKLearn, XGBoost and ensembling, exporting Python feature engineering scripts and model artifacts with a Studio link for inspection. JumpStart exposes a 150+ model model zoo with one‑click deploy to endpoints and ready‑to‑run notebooks that show inference code. SageMaker Studio runs MLOps pipelines (example pipeline completes in ~24 minutes), and Studio Lab offers free GPU sessions (≈4 hours) and CPU sessions (≈12 hours) for learning.
AI-written summary · about
chapters (7)
- 0:00Introduction to Sagemaker
- 2:37ML Stack and workflows
- 7:11Low-code and no-code tools
- 13:50Demo: Canvas and JumpStart
- 24:26SageMaker Studio overview
- 33:58Demo: Studio and MLOps
- 42:04Studio Lab and wrapping up
related talks
31mTrain and tune ML models to the highest accuracy using Amazon SageMaker
33mBuild, train, and deploy ML models using Amazon SageMaker
34mBuild, train and deploy Machine Learning models with Amazon SageMaker
31mBuild, train, and deploy ML models using Amazon SageMaker
21mAWS On Air ft. Amazon SageMaker Canvas
32mBuild a prediction ML model using public data
29mGet started with ML in minutes with Amazon SageMaker JumpStart
32mIntroduction to Amazon SageMaker Studio, a Complete IDE for ML