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 metrics

Select metrics - Amazon Web Services (AWS) Tutorial

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

Start my 1-month free trial Buy for my team

Select metrics

“

- [Instructor] One of the classic problems with Machine Learning is you have to figure out what good is. And how do you do this? Well, one of the ways you can do this is by selecting a metric to figure out, for your problem, what it is that you're trying to solve. Now, let's take a very simple scenario here, where you're trying to predict between cats and dogs. So if you have, let's say, 10 items here and there are some cats and there's some dogs in there, you want to be able to determine how many times you were able to correctly identify a dog, and this would be a true positive, and how many times you were correctly able to identify that it was not a dog, it was a cat and this would be the false negative. So you can see this is a relevant element. So it's this category. Now, in the alternate scenario here, if you're able to, in this scenario, say that these were dogs, but they were really cats, this would be a false…

Contents