Instance Property


The fraction of incorrectly labeled examples.


var classificationError: Double { get }


The classification error describes how many examples were incorrectly labeled divided by the total number of examples. Accuracy as a percentage is often a more intuitive way to understand this. You can calculate it as follows:

let accuracy = (1 - metrics.classificationError) * 100

See Also

Understanding the Model

var precisionRecall: MLDataTable

A data table listing the precision and recall percentages for each category.

var confusion: MLDataTable

A table comparing the actual and predicted labels for each classification category.