AI Next Step: ML Ops

(c)Copyright Elephant Scale

July 1, 2021

Overview

Machine Learning and AI represent a great opportunity. All too often, taking a Machine Learning prototype to production makes a difference between success and failure in the AI strategy of a company. This need is fulfilled by Machine Learning Engineers who apply the rules of DevOps to AI.

This course teaches how to take Machine Learning and AI and reduce it to practice. Thus, the name: ML Ops.

Audience

Developers, team leads, project managers

Skill Level

Introductory – Intermediate

Duration

Three days

Format

Lectures and hands-on labs (50% – 50%)

Prerequisites

  • Comfortable developing code in the target environment

Lab environment

  • Zero Install: There is no need to install software on students’ machines!
  • A lab environment in the cloud will be provided for students.

Students will need the following

  • A reasonably modern laptop with unrestricted connection to the Internet. Laptops with overly restrictive VPNs or firewalls may not work properly.
    • A checklist to verify connectivity will be provided
  • Chrome browser

Detailed outline

Rise of the Machine Learning Engineer and MLOps

  • What is MLOps
  • DevOps and MLOps
  • An MLOps Hierarchy of Needs
  • Implementing DevOps
  • DataOps and Data Engineering
  • Platform Automation
  • MLOps
  • Where Can You Deploy?
  • Conclusion

MLOps Foundations

  • Bash and the Linux Command Line
  • Cloud Shell Development Environments
  • Bash Shell and Commands
  • List Files
  • Run Commands
  • Files and Navigation
  • Input/Output
  • Configuration
  • Writing a Script
  • Cloud Computing Foundations & Building Blocks
  • Machine Learning Key Concepts
  • Build an MLOps Pipeline from Zero

MLOps For Containers And Edge Devices

  • Containers
  • Serving a trained model over HTTP
  • Edge Devices
  • Coral
  • Azure Percept
  • TFHub

Continuous Delivery For Machine Learning Models

  • Packaging for ML Models
  • Infrastructure as Code for Continuous Delivery of ML Models
  • Using Cloud Pipelines
  • Controlled Rollout of models
  • Testing techniques for Model Deployment

Monitoring And Logging

  • Introduction to Logging
  • Logging in Python
  • Modifying log levels
  • Logging different applications
  • Monitoring Drift with AWS SageMaker
  • Monitoring Drift with Azure ML

MLOps For Azure

  • Azure CLI and Python SDK
  • Authentication
  • Service Principal
  • Authenticating API Services
  • Compute Instances
  • Deploying
  • Registering Models
  • Versioning datasets
  • Deploying Models to a Compute Cluster
  • Configuring a Cluster
  • Deploying a Model
  • Azure ML Pipelines

Machine Learning Interoperability

  • Why interoperability is critical
  • ONNX: Open Neural Network Exchange
  • Convert PyTorch into ONNX
  • Convert TensorFlow into ONNX
  • Apple Core ML