Azure Machine Learning Studio Pipeline Creation

1.5 hours
  • 5 Learning Objectives

About this Hands-on Lab

In this hands-on lab, you work as a Machine Learning Operations Engineer for Avendador. During the lab, you will create and execute a machine learning pipeline using sample Microsoft data sources (**Restaurant Ratings** and **Restaurant Feature Data**).

Learning Objectives

Successfully complete this lab by achieving the following learning objectives:

Preparing the Environment
  1. Create a Standard_D2_V2 compute instance and start it.
  2. Create an Azure Machine Learning pipeline in Azure Machine Learning studio.
Ingest Data and Select Key Columns
  1. Add the sample datasets, Restaurant Ratings and Restaurant Feature Data, to the pipeline canvas.
  2. Select placeID and rating from the Restaurant Ratings data source.
  3. Select placeID, alcohol, dress_code, price, and Rambience from the Restaurant Feature Data source.
Transform Data Sources
  1. Join the data sources using placeID as key.
  2. Replace missing data in columns (placeID, rating, alcohol, dress_code, price, Rambience) with 0.
Split Data into Training and Test Data
  1. Split data using a 60/40 split.
    • 60% should go to a filter using Pearson correlation
    • 40% should be used as test
  2. Create a Pearson correlation feature selection using rating as a target column (select columns to transform and apply transformation).
  3. Create a Boosted Decision Tree Regression with the following settings:
    • Create trainer mode: SingleParameter
    • Maximum number of leaves per tree: 20
    • Minimum number of leaves per tree: 10
    • Learning rate: 0.2
    • Total number of trees constructed: 100
  4. Create Train Model using rating as label column.
Score and Evaluate
  1. Create Score Model activity.
  2. Create Evaluate Model activity.
  3. Submit Model.
  4. Evaluate Results.

Additional Resources

Lab Scenario

In this hands-on lab scenario, you are a Machine Learning Operations Engineer for Avendador. Avendador has data on Mexican restaurants and customer preferences. They have worked with a data scientist to select features and choose an algorithm. They would like you to build a machine learning pipeline to see if there is a correlation from the data.

To accomplish your goal, the following should be completed:

  • Prepare the environment.
  • Create a pipeline and ingest the data.
  • Prepare the data.
  • Perform feature selection and apply an algorithm.
  • Score and evaluate model performance.

Reference Links

Additional information on Pearson correlation found on the Azure Filter Based Feature Selection page.

What are Hands-on Labs

Hands-on Labs are real environments created by industry experts to help you learn. These environments help you gain knowledge and experience, practice without compromising your system, test without risk, destroy without fear, and let you learn from your mistakes. Hands-on Labs: practice your skills before delivering in the real world.

Sign In
Welcome Back!

Psst…this one if you’ve been moved to ACG!

Get Started
Who’s going to be learning?