Run Experiments and Train Models
John Ellis
0 h 50 m
Lecture Overview
To get results from Azure Machine Learning, you must run Experiments. In this course we learn what experiments are, how they relate to Models and how to create them. Furthermore, this course will show you how to obtain output and metrics from Experiments and even how to automate the process using Azure Machine Learning. 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
  • Be able to use the Azure Machine Learning Designer
  • Be able to construct and run scripts in a workspace environment
  • View and understand metrics output from Experiment runs
  • Be able to automate the execution of Experiments
Lecture Modules
This module is divided into 4 sections. First, we explore what a Model is in context of Azure Machine Learning. Next, we explore the concept of Pipelines in Azure Machine Learning. After this we will take another brief look at Azure Machine Learning designer which is part of the Enterprise subscription for Azure Machine Learning.
Finally, we will demonstrate how to create our first pipeline using the graphical designer interface in the Azure Machine Learning studio portal.
This module is divided into three sections with a demonstration. We first will explore the concept behind Experiments and understand how they relate to models.
Next, we have a review of Datastores and Datasets. Third, we introduce a new concept called Estimators and discuss how they are abstraction upon models and experiments. Finally, we will demonstrate how to bring this all together in order to create an Experiment using the Python SDK.
This module covers two short but important topics for Azure Machine Learning: Logging and Experiment output. We will take a look at each of these before concluding with a demonstration of these two subjects.
This module is all about how we automate our tasks by creating pipelines with the python SDK. We will first review what the components of a pipeline are, how to create them with the SDK, then we will discuss how to pass data from one step of the pipeline to another. Finally we will explore how to run and monitor our pipeline from the SDK.
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