Model Evaluation - Regression Models

This guide provides a comprehensive walkthrough for evaluating regression models using Azure AI and Machine Learning Studio, making it an essential resource for data scientists and machine learning practitioners.

Hafeez Baig
18 steps
2 minutes

Azure

This lab is a continuation of the previous one, Please refer to it before proceeding - <https://scribehow.com/viewer/Creating_Azure_Machine_Learning_Pipeline__fx5Ai4fzQbypi3O7sqxsig>

On the **Azure AI** page, click on the **Designer** option from the left bar

**What is Designer in Azure AI?** **Designer** in Azure AI (specifically in Azure Machine Learning) is a **drag-and-drop visual interface** that allows users to build, train, and deploy machine learning models without writing code. It enables: - Easy assembly of ML pipelines using prebuilt modules - Data preparation, training, and evaluation visually - Rapid prototyping and experimentation for both beginners and experts Designer simplifies the end-to-end ML workflow in a user-friendly, low-code environment.

**Designer** page will open, scroll the page to the **Pipelines** section and click on the **Pipeline jobs** tab

Click on the **custom-ml-experiment** under **Experiments** column

**custom-ml-experiment** page will open, click on the **custom-ml-pipeline** name

**custom-ml-pipeline** page will open, here you can see the pipeline is **Running**

On the right side click on the **Job overview** option

**Pipeline job overview** wizard will open on the right side, here you can see the **Properties**

Click on the **Child jobs** tab, here you can view the jobs running

Click on the **Output + logs** tab, here you can view the output logs

On the pipeline wizard, Click on the **Evaluate Model**, select the **Preview data** and click on the **Evaluate results** option

**Evaluate results** wizard will open on the right side, click on the expand icon

Here you can see the Rows, Columns and its evaluation details

Click on the **Publish** button on the top middle

**Set up published pipeline** wizard will open, click on the **Create new** radio button for **Pipeline Endpoint**

Give the **New PipelineEndpoint name** as "**custom-ml-pipeline**" and click on the **Publish** button

On the top left side you can see **Published succeeded** text, click on the **custom-ml-pipeline** link

**Pipeline endpoint overview** wizard will open on the right side, here you can see the overview and **REST endpoint**

Scroll the wizard and here you can see the **REST endpoint documentation** and its details

Congratulations on completing this lab and exploring model evaluation pipeline!