Deploy and Consume Models
Lecture
John Ellis
Intermediate
0 h 36 m
2020-09-18
Lecture Overview
Now you have a good, trained model. How can you use it? In this course we explore the various ways that models can be deployed for making single predictions as well as batch predictions. You will learn the various types of deployment environments that exist and how to choose compute for those. This course will also teach you how to convert your existing training pipelines into deployable inferencing pipelines. This course covers one of the objective areas that help you in preparing for the DP-100: Designing and Implementing a Data Science Solution on Azure exam.

Related Learning Path(s):
DP-100: Designing and Implementing an Azure Data Science Solution on Azure
Objectives
  • Understand compute targets for deploying services
  • Be able to deploy a model as a real-time inferencing webservice
  • Be able to deploy a pipeline for batch inferencing
  • Be able to deploy pipelines via the Azure Machine Learning Designer
Lecture Modules
In this module we will learn about production compute targets for deploying inferencing endpoints and then we will see a demonstration that shows the creation of compute targets for both real-time and batch inferencing services.
This module is divided into three sections with one demonstration. First, we will explore what real-time inferencing is. Then, we discuss how models can be deployed to become a web service endpoint that can serve up single predictions on new data occurrences. Third, we will cover how to troubleshoot a real-time inferencing webservice. And finally we will view a demonstration in which I create and deploy a simple real-time inferencing webservice.
This module is divided into two sections with a demonstration of the content at the end. First, we will talk about batch inferencing. Then, we will go through the steps required to deploy our batch inferencing service. Finally we will view a demonstration of these concepts in action.
This module will explain how a visual training pipeline can be converted into a real-time inferencing pipeline using the Azure Machine Learning studio designer. We will then conclude with a demonstration that creates the pipeline, converts it, deploys it and tests the output.
Try Risk Free

Start a free trial

Skill Me Up subscriptions include unlimited access to on-demand courses with live lab lab environments with our Real Time Labs feature for hands-on lab access.

Subscription Benefits
  • Access to Real Time Lab environments and lab guides
  • Course Completion Certificates when you pass assessments
  • MUCH MORE!