Argo Workflow Intermediate

Teaching methodology

Prerequisites

• Basic knowledge of Argo Workflow or participation in the DSAI105 introductory course.
• Understanding the fundamental principles of machine learning.
• Familiar with process automation and orchestration concepts.

1. Introduction to MLOps Concepts: This module provides an overview of the fundamental concepts of MLOps, which focuses on automating workflows for developing, testing, and releasing machine models learning. Students learn the importance of integrating development and operations processes to improve the efficiency and scalability of machine learning operations.
2. Argo Essential Recap: This module introduces the basic concepts and functionality of Argo Workflow, an open-source workflow orchestration engine designed for performing complex operations in production environments.
3. Integrating Argo Workflow into an MLOps environment: Students learn best practices and approaches for integrating Argo Workflow into an MLOps environment. Through case studies and practical examples, students gain skills to implement and manage machine learning workflows in an efficient and scalable way.
4. Using workflow templates throughout the model lifecycle: This module illustrates how to use workflow templates to automate and standardize the lifecycle of your machine learning model, from ‘assessment and release training.
5. Implementing Event-Driven Pipelines: Students learn how to design and implement event-driven machine learning pipelines, which dynamically respond to changes in the environment or input data.
6. Monitoring Workflows Using Related Tools: This module introduces the tools and techniques to monitor the execution of workflows, collecting metrics and reporting any problems or anomalies.
7. Workflow management using related tools: Students gain skills for managing and optimizing workflows using related tools, such as queue management systems and container orchestration.< /li>
8. Performance Optimization of Machine Learning Workflows: This module focuses on optimizing the performance of machine learning workflows, including parallelization, large-scale deployment, and resource optimization.

• Use Argo Workflow in an advanced way to automate processes related to machine learning.
• Understand the fundamental principles of MLOps and the role of Argo Workflow in this context.
• Implement and manage complex workflows for developing, testing, and releasing machine learning models.
• Integrate Argo Workflow into an operational machine learning infrastructure following best practices.
• Monitor and optimize workflows to maximize the performance of machine learning models.

• Internet connection
• Web browser, Google Chrome
• Zoom

• Instructor: Italian
• Workshops: English
• Slides: English