5 Customizing the search space by creating AutoML pipelines

 

This chapter covers

  • Understanding an AutoML pipeline
  • Customizing sequential and graph-structured AutoML pipelines
  • Automated hyperparameter tuning and model selection with customized AutoML pipelines
  • Customizing the AutoML blocks in the AutoML pipeline

In chapter 4, we solved a variety of problems with AutoKeras without customizing the search space. To recap, in AutoML, a search space is a pool of models with specific hyperparameter values that potentially can be built and selected by the tuning algorithm. In practice, you may want to use a specific ML algorithm or data preprocessing method to solve a problem, such as an MLP for a regression task. Designing and tuning a particular ML component requires tailoring the search space, tuning only the relevant hyperparameters while fixing some others.

5.1 Working with sequential AutoML pipelines

 
 

5.2 Creating a sequential AutoML pipeline for automated hyperparameter tuning

 
 

5.2.1 Tuning MLPs for structured data regression

 
 

5.2.2 Tuning CNNs for image classification

 
 
 

5.3 Automated pipeline search with hyperblocks

 

5.3.1 Automated model selection for image classification

 
 
 

5.3.2 Automated selection of image preprocessing methods

 
 
 
 

5.4 Designing a graph-structured AutoML pipeline

 
 

5.5 Designing custom AutoML blocks

 

5.5.1 Tuning MLPs with a custom MLP block

 

5.5.2 Designing a hyperblock for model selection

 
 
 

Summary

 
 
 
sitemap

Unable to load book!

The book could not be loaded.

(try again in a couple of minutes)

manning.com homepage
test yourself with a liveTest