8221  Reviews star_rate star_rate star_rate star_rate star_half

Introduction to AIOps

This in-person or online AIOps (Artificial Intelligence Operations) training class teaches attendees how to successfully deploy AI and Data Science systems at scale. Students learn how to break down...

Read More
Course Code DATA-104
Duration 3 days
Available Formats Classroom

This in-person or online AIOps (Artificial Intelligence Operations) training class teaches attendees how to successfully deploy AI and Data Science systems at scale. Students learn how to break down a system or pipeline into functional components, scale different types of processes, and adjust for various types of Big Data requirements.

Skills Gained

  • Understand Data Science, including the Data Science Life Cycle
  • Understand the types of applications of Machine Learning
  • Understand what AIOps is and how it builds on top of traditional DevOps in a cloud environment
  • Understand considerations for infrastructures and topologies, including on-prem, hybrid, and micro-services variations
  • Understand the need for model explainability, both from a technical and business perspective
  • Use AutoML and other automation technologies (using AWS examples)
  • Work with intermediate data within a pipeline

Prerequisites

All students must have an analytics and/or Python background. Familiarity with AWS or other cloud environments is strongly encouraged. Students should have a familiarity with how data science and machine learning are used, at least from a business or product perspective. A general understanding of cloud DevOps is also strongly encouraged.

Course Details

Training Materials

All AIOps training students receive comprehensive courseware.

Software Requirements

Students should have Python 3 installed with the ability to install other packages or programs (i.e., Admin Access) on their laptops. Anaconda with Python 3 is strongly recommended over the python.org installation.

Outline

  • Introduction
  • Data Science
    • Overview
    • Machine Learning
    • Asking the Right Questions
    • Artificial Intelligence: ML + Knowledge
    • The Data Science Pipeline
    • The Data Science Life Cycle
    • Data Science and AIOps
  • Machine Learning
    • ML for Analytics
    • ML for Prediction
    • ML for Regression
    • Scaling ML
  • AIOps
    • The Need for AIOps
    • The IT Operations Management Cycle
  • The Five Dimensions of AIOps
    • Data Set Selection
    • Pattern Discovery
    • Inference
    • Communications
    • Automation
  • Infrastructure and Topologies
    • Cloud, On-Premise, and Hybrid Cloud
    • Micro Services
    • Scaling
    • Cost Projections
    • The Failure of Traditional ITOM Technologies
    • Industry Examples
  • Model Explainability
    • Why are we getting these predictions?
    • Model reductions for explainability
    • Other trending techniques and solutions
  • Working with the Components
    • AWS
    • Data (AWS-S3)
      • Compute (EC2, deploying an API, loading data from S3)
      • AWS ML (an ML API endpoint)
    • Working Locally
      • Data Wrangler
      • Saving Intermediate Datasets
      • Flask API’s
      • Tableau for a Front End
  • Practical Exercise
    • Build a simple analytics app
    • Connect to data via API
    • Build a Data Science Pipeline as a middle layer
    • Connect to UI/front end (Tableau)
  • Emerging Trends
    • Emerging Technologies
      • Micro Services
      • Auto ML
      • NLP Trends and Techniques
      • Graph Databases and Network Graph Analysis
  • Conclusion