.svg)
.svg)
.svg)
Machine Learning Engineer
ML data engineers engineers, engineers engineers and developers responsible for operationalizing ML models will learn to overcome the challenges of handover between data engineers, data scientists, software developers and operations staff through the use of tools, automation, processes and teamwork. At the end of the course, you will be able to move from learning to action by creating an MLOps action plan for your organization.
Day 1
Module 0: Welcome
- Introduction to the course
Module 1: Introduction to MLOps
- Machine learning operations
- Objectives of MLOps
- Communication
- From DevOps to MLOps
- ML workflow
- Scope
- MLOps view of the ML workflow
- MLOps cases
Module 2: MLOps development
- Introduction to creating, training and evaluating machine learning models
- MLOps security
- Automate
- Apache Airflow
- Kubernetes integration for MLOps
- Amazon SageMaker for MLOps
- Exercise: Bring your own algorithm into an MLOps pipeline
- Demo: Amazon SageMaker
- Introduction to creating, training and evaluating machine learning models
- Exercise: Code and deploy your ML model with AWS CodeBuild
- Activity: MLOps action plan workbook
Day 2
Module 3: MLOps provision
- Introduction to the provisioning processes
- Model packaging
- Inference
- Exercise: Using the model in production
- SageMaker production variants
- Strategies for use
- Operation at the border
- Exercise: Performing A/B tests
- Activity: MLOps action plan workbook
Day 3
Module 4: Model monitoring and operation
- Exercise: Troubleshooting in your pipeline
- The importance of monitoring
- Monitoring by design
- Exercise: Monitor your ML model
- Man in the loop
- Amazon SageMaker model monitoring
- Demo: Amazon SageMaker Pipelines, Model Monitor, Model Registry and Feature
- Save
- Solving the problem(s)
- Activity: MLOps action plan workbook
Module 5: Follow-up
- Repetition of the course
- Activity: MLOps action plan workbook
- Follow-up
- Deploy your own models in the AWS Cloud
- Automate workflows to create, train, test and deploy ML models
- The different deployment strategies for the implementation of ML models in production
- Monitoring of data and concept deviation that could affect forecasting and alignment with business expectations
This course consists of training training and is led by a trainer who supervises the participants live. Theory and practice are taught with live demonstrations and practical exercises. The video conferencing software Zoom is used.
This course is aimed at the following job roles:
- DevOps
- Machine learning & AI
.svg)
