Machine Learning Operations (MLOps)

In this one-day workshop, you will learn how to operationalize Machine Learning models using popular open-source tools (Kedro), and deploy it using cloud computing (Google Cloud Platform / Vertex AI).

During the course, we simulate real-world end-to-end scenarios of building a Machine Learning pipeline to train a model and deploy it in the cloud environment. We’ll walk through the practical use cases of MLOps for creating reproducible, scalable, and modular data science code. Next, we’ll propose a solution for running pipelines on Google Cloud Platform, leveraging managed and serverless services. All exercises will be done using either a local docker environment or GCP account.

   Target Audience

Data scientists and DevOps who are interested in implementing MLOps best practices, and build Machine Learning pipelines.

    Requirements

Some experience coding in Python, a basic understanding of cloud computing, and machine learning concepts

    Participant’s ROI

• Practical knowledge of building Machine Learning pipelines using Kedro
• Hands-on experience with building Machine Learning platform with Kubeflow Pipelines
• Tips about real-world applications and best practices.

    Training Materials

All participants will get training materials in the form of PDF files containing slides with theory and an exercise manual with a detailed description of all exercises. During the workshops, the exercises can be done using either a local docker environment or within your IDE.

    Time Box

This is a one-day event (9:00-16:00), and there will be some breaks between sessions.

    Agenda

Session #1 - Introduction to Machine Learning Operations (MLOps)

• Introduction and key concepts
• MLOps components
• The challenges of deploying and maintaining Machine Learning models in production
• The Machine Learning model lifecycle

Session #2 - Kedro - a framework to structure your ML pipeline

• Create reproducible, maintainable, and modular data science code
• Build your Machine Learning pipeline
• Hands-on exercises

Session #3 - Running Machine Learning pipelines in on the cloud environment

• Introduction and key concepts
• Example of Google Cloud Platform / Vertex AI Pipelines (serverless) deployment
• Hands-on exercises

Session #4 - Building infrastructure for your Machine Learning platform

• Overview of MLOps Frameworks landscape, and reference architectures
• Write once - run everywhere: showcase of Kedro pipelines run on various clouds with Kedro plugins (Google Cloud Vertex AI, Amazon Sagemaker, Microsoft AzureML)

Session #5 - Summary and wrap-up

   Participants limit
18 participants

   Keywords: MLOps, Kubeflow, Kubeflow Pipelines, Kedro, Google Cloud Platform

    Session leader: