Overview

The focus of this course is on the tools and services available in IBM Cloud Pak for Data that can be used to build, test, and deploy machine learning models. It takes the Data Scientist or Business Analyst on a journey from the creation of several machine learning models to its deployment and testing. Various tools and services, as well as programming and graphical user interfaces, are used in the process. The course ends with the sharing of assets on GitHub, and a brief discussion on governance and stewardship.

Audience

The course is designed for Data Scientists and Business Analysts. It caters to aspiring or practicing Data Scientists, and Business Analysts with prior knowledge of Data Science and Machine Learning, but lacking familiarity with IBM tools and IBM Cloud Pak for Data. Rather than teaching data science and machine learning, the course focuses on demonstrating how IBM tools and services can address significant business challenges, targeting professionals at the associate level and beyond.

Prerequisites

Before taking this course, you should have:

• Knowledge of Data Science
• Knowledge of Machine Learning
• Experience with the Python programming language

Objective

• Define a solution to a business problem using tools and frameworks from IBM Cloud Pak for Data
• Demonstrate how the AI lifecycle can be automated by building a rapid prototype using AutoAI
• Build, train, and deploy a machine learning model with the tools and services available in Cloud Pak for Data
• Implement GitHub Integration and team collaboration in Cloud Pak for Data

Course Outline

Introduction

• Discuss the four steps that make up the AI Ladder
• Explain the AI Lifecycle and the different personas that are involved in an AI project
• Explain the components that comprise a project

Rapid prototyping with AutoAI

• Describe the benefits of AutoAI for rapid prototyping
• Discuss the steps needed to run a successful AutoAI experiment
• Interpret and discuss the results from an AutoAI experiment

Creating, testing, and deploying machine learning models

• Discuss various ways of building a machine learning model in Watson Studio by using programming and visual interfaces
• Discuss the contents of the model repository
• Describe the various deployment methods for AI models

Governance, integration, and collaboration

• Describe the capabilities and features to support governance of your data
• Describe the team collaboration features