From the course: AWS Certified Machine Learning - Specialty (MLS-C01) Cert Prep: 3 Modeling

Unlock this course with a free trial

Join today to access over 23,100 courses taught by industry experts.

Select models

Select models

- [Instructor] How do you select the correct, appropriate model for the problem you're solving? There is no perfect solution for this, and there are many different techniques that you can think about. But one that's a very interesting technique is to first use an AutoML tool. Now, you can't just throw an AutoML tool at a problem without selecting the problem type first. You need to pick between a regression, let's say, or a classification. Let's say we picked a classification. You can then say, I'd like the AutoML system to pick from a variety of models and a variety of hyperparameters, a bunch of different experiments. And once it goes through, and it tries let's say Xgboost, Logistic regression, tries a bunch of other classification models with lots of different hyperparameters, maybe using, you know, a grid search to do that. All of those experiments will show up in SageMaker experiment tracking, or another tool…

Contents