From the course: Machine Learning with SageMaker by Pearson

Unlock this course with a free trial

Join today to access over 25,200 courses taught by industry experts.

Model over/underfitting demonstration

Model over/underfitting demonstration - Amazon SageMaker Tutorial

From the course: Machine Learning with SageMaker by Pearson

Start my 1-month free trial Buy for my team

Model over/underfitting demonstration

“

In this demo, we're going to be looking at underfitting and overfitting and how the manipulation of hyperparameters can result in overfitting and underfitting as well, not supplying enough data or having unbalanced data. And the way I'm going to execute this demo, I'm not going to use the studio feature of SageMaker. Remember that once you start using Studio or Canvas, you're going to start paying quite a bit more. For example, as I record this, to run Studio 24 hours a day is about $45 per day. And I multiply that by 30, you can see how your price can go up quite a bit. So I just want to demonstrate to you that you don't have to use Studio to use SageMaker. So over here in SageMaker, as we see on the screen right here, I don't even have a domain created right now. But I am still going to train a model and I'm going to deploy that model and I'm going to run an inference against that model. I'm actually going to train three different models. I'm going to create an underfitted model, a…

Contents