Evaluate the performance of a prediction.

Description

Wraps the problem-specific evaluation functions by calling evaluate_problem. This wrapper is desirable, as it can perform the extraction of the holdout set (observations)

Usage

1
evaluate(prediction, data, test, ...)

Arguments

prediction

A vector of predictions for each row in the holdout set

data

The data list containing train and holdout data sets

test

The test object being evaluated

...

Extra arguments to evaluate

Value

An object of class 'evaluation', which contains a list of performance measures and a test object.


Questions? Problems? Suggestions? or email at ian@mutexlabs.com.

All documentation is copyright its authors; we didn't write any of that.