From the course: Exploring Data Science with .NET using Polyglot Notebooks & ML.NET

Unlock this course with a free trial

Join today to access over 24,000 courses taught by industry experts.

Evaluating classification models

Evaluating classification models

From the course: Exploring Data Science with .NET using Polyglot Notebooks & ML.NET

Start my 1-month free trial Buy for my team

Evaluating classification models

“

- [Instructor] Now that we used ML.NET and AutoML to train a series of binary classification models, let's take a look at how good these models are. Here I have a run detail of binary classification metrics, representing the best run that AutoML found. My results object also has a run details collection that gives me a list of all the different models it evaluated, but usually you'll want to start by looking at the best run. Now on this best run, we have a number of metrics associated with our binary classification experiment. I'm going to go in here and I'm going to say bestRun.validationmetrics, and we'll take a look at what's rendered. So here we have a handful of different metrics associated with our model. We'll come back to these in a moment because most of these derive from something called the confusion matrix. And ML.NET gives us a really nice way of seeing a confusion matrix. I'm going to add a new cell here. I'm going to say bestRun.validationmetrics confusion matrix, and…

Contents