Description
In this course, you will learn :
- A feature store is used to store and manage machine learning features.
- Models can be debugged, profiled, tuned, and evaluated while data lineage and model artefacts are tracked.
Syllabus :
1. Feature Engineering and Feature Store
- Introduction to Feature Engineering
- Feature Engineering Steps
- Feature Engineering Pipeline
- BERT: Bidirectional Encoder Representations from Transformers
- BERT: Example
- Feature Engineering: At scale with Amazon SageMaker Processing Jobs
- Feature Store
- Amazon SageMaker Feature Store
2. Train, Debug, and Profile a Machine Learning Model
- Pre-trained models
- Pre-trained BERT models
- Train a custom model with Amazon SageMaker
- Debug and profile models
- Debug and Profile Models with Amazon SageMaker Debugger
3. Deploy End-To-End Machine Learning pipelines
- Machine Learning Operations (MLOps)
- Creating Machine Learning Pipelines
- Model Lineage & Artifact Tracking
- Machine Learning Pipelines with Amazon SageMaker Pipelines
- Machine Learning Pipelines with Amazon SageMaker Projects
- Amazon SageMaker Projects Demo