fr
en
Amazon SageMaker Studio for Data Scientists
Your AI use cases need suitable services and evaluated results. With AWS, connect data, models and integration to understand the technical options more clearly. Develop practical guidelines for designing experiments and assessing implementation conditions.
- Duration
- 3 days 21 hours
- Code
- AWS15FR Code
Presentation
Amazon SageMaker Studio helps data scientists prepare, build, train, deploy and monitor machine learning (ML) models quickly. It brings together a wide range of capabilities designed specifically for ML. This unified platform enables data scientists to improve productivity at every stage of the ML lifecycle by integrating tools such as Amazon CodeWhisperer and Amazon CodeGuru Security analysis extensions.
In this advanced course, you will prepare to master the tools and services that make up SageMaker Studio. You will explore key concepts in depth, including large-scale data processing with SageMaker Data Wrangler and Amazon EMR, and model development with SageMaker Experiments and SageMaker Debugger. You will also cover model deployment and monitoring with SageMaker Model Registry and Amazon SageMaker Model Monitor. Through demonstrations and numerous hands-on workshops, you will learn to optimise every phase of your ML projects.
By the end of this 3-day programme, you will have developed the expertise to accelerate the preparation, building, training, deployment and monitoring of your ML solutions using Amazon SageMaker Studio.
As a premium training partner (ATP) authorised by Amazon Web Services, Oo2 offers skills development and certification training that meets the organisation's rigorous quality standards.
Objectives
By the end of this Amazon SageMaker Studio course, you will be able to:
- master data preparation and analysis at different scales with services such as SageMaker Data Wrangler, Amazon EMR and AWS Glue;
- design and manage ML model development using features such as built-in algorithms, custom scripts and experiment tracking with SageMaker Experiments;
- deploy ML models efficiently for inference and optimise their performance using scaling and testing strategies;
- establish and manage continuous model monitoring to detect and address drift and performance issues;
- manage SageMaker Studio resources, including costs and updates.
Program
Module 1: configuring SageMaker Studio
- Exploring JupyterLab extensions in SageMaker Studio.
- Introduction to the SageMaker user interface (demo).
Module 2: processing data
- Using SageMaker Data Wrangler for data processing.
- Using Amazon EMR to analyse and prepare data at scale.
- Using AWS Glue interactive sessions.
- Using SageMaker Processing with custom scripts.
- Managing features for feature engineering with SageMaker Feature Store.
Labs:
- Analyse and prepare data with Amazon SageMaker Data Wrangler.
- Analyse and prepare data at scale with Amazon EMR.
- Process data using Amazon SageMaker Processing and the SageMaker Python SDK.
- Extract features from raw data using SageMaker Feature Store.
Module 3: developing models
- Understanding training jobs.
- Using built-in algorithms.
- Creating a custom script.
- Creating a custom container.
- Tracking model training and tuning iterations with SageMaker Experiments.
- Analysis, detection and alert configuration with SageMaker Debugger.
- Automatic model tuning.
- Introduction to ML automation with SageMaker Autopilot (demo).
- Detecting bias.
- Using SageMaker JumpStart.
Labs:
- Track training iterations and tune models with SageMaker Experiments.
- Analyse, detect and configure alerts using SageMaker Debugger.
- Use SageMaker Clarify for bias and explainability.
Module 4: deployment and inference
- Registering models with SageMaker Model Registry.
- Orchestrating ML workflows with SageMaker Pipelines.
- SageMaker model inference options.
- Scaling inference.
- Testing, performance and optimisation strategies.
Labs:
- Use SageMaker Pipelines and SageMaker Model Registry with SageMaker Studio.
- Deploy a machine learning model for inference with SageMaker Studio.
Module 5: monitoring models
- Monitoring models with Amazon SageMaker Model Monitor.
- Case study discussion.
- Introduction to model monitoring (demo).
Module 6: managing SageMaker Studio resources and updates
- Managing accumulated costs and shutting down resources.
- Performing updates.
Challenge labs:
- Set up the working environment.
- Analyse and prepare the dataset with SageMaker Data Wrangler.
- Create feature groups in SageMaker Feature Store.
- Perform and manage model training and tuning using SageMaker Experiments.
- Use SageMaker Debugger to optimise training and model performance (optional).
- Evaluate the model for bias using SageMaker Clarify.
- Perform batch predictions using a model endpoint.
- Automate the entire model development process using SageMaker Pipeline (optional).
Audience
This course is intended for:
- Experienced data scientists who are comfortable with machine learning and deep learning fundamentals.
Prerequisites
The following prerequisites are recommended for this AWS course:
- experience with ML frameworks;
- Python programming experience;
- at least 1 year of experience as a data scientist responsible for training, tuning and deploying models;
- completion of AWS Technical Essentials.
Teaching and assessment methods
- Initial skills assessment
- Training materials provided to participants
- Continuous assessment throughout the course
- End-of-course feedback questionnaire
- Combination of theory and practical application
- Attendance records
- Post-course follow-up evaluation
- Practical exercises
Course highlights
- Expert AWS-certified instructors: benefit from recognised, AWS-certified instructors with in-depth knowledge of ML practices and AWS tools and services for model development and deployment.
- Interactive hands-on learning: master machine learning tools and techniques on AWS through practical demonstrations and workshops. Prepare to address real-world challenges in data preparation, model training and tuning, and cloud monitoring.
- Key skills development: the course content is carefully designed to help you acquire essential skills in designing and implementing ML solutions, automating model development and deployment processes, and managing AWS resources effectively.
Dates and sessions
Choose the date and delivery format that suit you.
No upcoming sessions are currently available.
AWS, Amazon SageMaker Studio, Amazon CodeWhisperer and other AWS marks are registered trademarks of Amazon.com, Inc. or its affiliates.