Understand what MLOps is , how to get started and best practices using MLOps
Perform Continuous Integration for Python ML Projects
Use the AWS Cloud for MLOps development
Create Containerized workflows for MLOps
Create Flask and CLI Services for Python ML Projects
1. Getting Started with MLOps
Poll: Experience Level With MLOps?
Poll: Experience Level With Cloud Computing?
- What is MLOps and how to get started
- Why Cloud-based development environments for MLOps?
Exercise: Setup AWS Cloud9 Environment
Exercise: Setup Github and Git
- Why Cloud Based continuous integration
Exercise: Setup AWS Code Build
2. Building containerized MLOps command-line tool
Poll: Experience level with containers?
Poll: Experience level with command-line tools?
- Docker Overview
- Why Docker Containers vs Virtual Machines?
Exercise: Use a Docker Container from Docker Hub
Exercise: Extend a Docker Container
Exercise: Build a Python click command-line tool in a container
- Common Issues Running a Docker Container
3. Build Containerized ML Web Microservice Applications
Poll: Experience level with running containers?
Poll: Experience level with container registries?
- Flask Microservice Overview
Exercise: Build a Flask Docker sklearn prediction container in AWS Cloud9
Exercise: Run a Flask Docker sklearn prediction container in AWS Cloud9
Exercise: Verify inference response from Flask application using utilities you build yourself.
4. Continuous Delivery Containerized App
Poll: Experience level with building containers automatically?
Exercise: Deploy a Docker sklearn prediction container to Docker Hub
Exercise: Deploy a Docker sklearn prediction container to Amazon Container Registry
Exercise: Deploy Flask ML microservice container via AWS App Runner in a CaaS (Container as a Service) Workflow
Linux and Cloud knowledge
Access to live training and QA session with the Instructor
Access to the on-demand recording
Certificate of completion