Azure Machine Learning Studio for The Non-Data Scientist: Learn how to create experiments, operationalize them using Excel and Angular .Net Core applications, and create retraining programs by Michael Washington
English | 18 Jul. 2017 | ISBN: 1548871125 | ASIN: B0742M7XM7 | 154 Pages | AZW3 | 5.63 MB
English | 18 Jul. 2017 | ISBN: 1548871125 | ASIN: B0742M7XM7 | 154 Pages | AZW3 | 5.63 MB
Creating predictive models is no longer relegated to data scientists when you use tools such as the Microsoft Azure Machine Learning Studio.
Azure Machine Learning Studio is a web browser-based application that allows you to create and deploy predictive models as web services that can be consumed by custom applications and other tools such as Microsoft Excel.
With this book, you will learn how to create predictive experiments, operationalize them using Excel and Angular .Net Core applications, and create retraining programs to improve predictive results.
Table of Contents
Chapter 1: The Author is Not a Data Scientist
Why Do We Need Predictive Modeling?
An Introduction to Get You Started
Chapter 2: An End-To-End Azure Machine Learning Studio Application
Create an Azure Machine Learning Workspace
Create An Experiment
Select Columns
Split Data
Train The Model
Score The Model
Evaluate The Model
Create A Predictive Web Service
Consume The Model Using Excel
Chapter 3: An Angular 2 .Net Core Application Consuming an Azure Machine Learning Model
The Application
Creating The Application
Create The .Net Core Application
Add PrimeNG
Add The Database
Create Code To Call Azure Machine Learning Web Service
Create The Angular Application
Saving Data
Viewing Data
Chapter 4: Retraining an Azure Machine Learning Application
The Retraining Process
Prepare The Training Data
Set-up An Azure Storage Account
Create The Batch Retraining Program
Get Required Values
Add A New Endpoint And Patch It
Consume The New Endpoint