Course Overview
Learn to boost productivity at every step of the ML lifecycle with Amazon SageMaker Studio for Data Scientists from an expert AWS instructor. The three-day, advanced level course helps experienced data scientists build, train, and deploy ML models for any use case with fully managed infrastructure, tools, and workflows to reduce training time from hours to minutes with optimized infrastructure. This course includes presentations, demonstrations, discussions, labs, and at the end of the course, you’ll practice building an end-to-end tabular data ML project using SageMaker Studio and the SageMaker Python SDK.
Course Content
- Amazon SageMaker Setup and Navigation
- Data Processing
- Model Development
- Deployment and Inference
- Monitoring
- Managing SageMaker Studio Resources and Updates
- Capstone
Who should attend
- Experienced data scientists who are proficient in ML and deep learning fundamentals.
- Relevant experience includes using ML frameworks, Python programming, and the process of building, training, tuning, and deploying models.
Prerequisites
We recommend that all students complete the following AWS course prior to attending this course:
We recommend students who are not experienced data scientists complete the following two courses followed by 1-year on-the-job experience building models prior to taking this course:
- !
- Deep Learning on AWS
Course Objectives
- Accelerate the preparation, building, training, deployment, and monitoring of ML solutions by using Amazon SageMaker Studio
- Use the tools that are part of SageMaker Studio to improve productivity at every step of the ML lifecycle
- And much more
Outline: Amazon SageMaker Studio for Data Scientists (ASSDS)
Day 1
Module 1: Amazon SageMaker Studio Setup
- JupyterLab Extensions in SageMaker Studio
- Demonstration: SageMaker user interface demo
Module 2: Data Processing
- Using SageMaker Data Wrangler for data processing
- Hands-On Lab: Analyze and prepare data using Amazon SageMaker Data Wrangler
- Using Amazon EMR
- Hands-On Lab: Analyze and prepare data at scale using Amazon EMR
- Using AWS Glue interactive sessions
- Using SageMaker Processing with custom scripts
- Hands-On Lab: Data processing using Amazon SageMaker Processing and SageMaker Python SDK
- SageMaker Feature Store
- Hands-On Lab: Feature engineering using SageMaker Feature Store
Module 3: Model Development
- SageMaker training jobs
- Built-in algorithms
- Bring your own script
- Bring your own container
- SageMaker Experiments
- Hands-On Lab: Using SageMaker Experiments to Track Iterations of Training and Tuning Models
Day 2
Module 3: Model Development (continued)
- SageMaker Debugger
- Hands-On Lab: Analyzing, Detecting, and Setting Alerts Using SageMaker Debugger
- Automatic model tuning
- SageMaker Autopilot: Automated ML
- Demonstration: SageMaker Autopilot
- Bias detection
- Hands-On Lab: Using SageMaker Clarify for Bias and Explainability
- SageMaker Jumpstart
Module 4: Deployment and Inference
- SageMaker Model Registry
- SageMaker Pipelines
- Hands-On Lab: Using SageMaker Pipelines and SageMaker Model Registry with SageMaker Studio
- SageMaker model inference options
- Scaling
- Testing strategies, performance, and optimization
- Hands-On Lab: Inferencing with SageMaker Studio
Module 5: Monitoring
- Amazon SageMaker Model Monitor
- Discussion: Case study
- Demonstration: Model Monitoring
Day 3
Module 6: Managing SageMaker Studio Resources and Updates
- Accrued cost and shutting down
- Updates
Capstone
- Environment setup
- Challenge 1: Analyze and prepare the dataset with SageMaker Data Wrangler
- Challenge 2: Create feature groups in SageMaker Feature Store
- Challenge 3: Perform and manage model training and tuning using SageMaker Experiments
- (Optional) Challenge 4: Use SageMaker Debugger for training performance and model optimization
- Challenge 5: Evaluate the model for bias using SageMaker Clarify
- Challenge 6: Perform batch predictions using model endpoint
- (Optional) Challenge 7: Automate full model development process using SageMaker Pipeline