Tuning Hyperparameters in Azure Machine Learning

Difficulty: Intermediate
Duration: 1 minute and 22 seconds
Students: 115
Rating: 5/5

In Tuning Hyperparameters in Azure Machine Learning, we see how to use hyperparameters to arrive at an optimal model solution. The training process is somewhat trial and error, so we start by looking at hyperparameter value selection. Then, we see how to run multiple trials with those values and, finally, how to get the best result without exhaustively trialling all possible hyperparameter values.

Learning Objectives

  • Hyperparameters overview
  • Sweep jobs
  • Early termination strategies

Intended Audience

Students preparing for the DP-100: Designing and Implementing a Data Science Solution on Azure exam and those who want to learn how to optimize model training with input or hyperparameter variables.

Prerequisites

Familiarity with data science concepts such as:

  • Models
  • Statistical analysis
  • Command jobs

It will be helpful to have taken the Running and Monitoring Training Scripts in Azure Machine Learning lesson.

Covered Topics