[source]

Confusion Matrix#

A Confusion Matrix is a table that visualizes the true positives, false positives, true negatives, and false negatives of a classification experiment. The name stems from the fact that the matrix makes it clear to see if the estimator is confusing any two classes.

Estimator Compatibility: Classifier, Anomaly Detector

Parameters#

This report does not have any parameters.

Example#

use Rubix\ML\CrossValidation\Reports\ConfusionMatrix;

// Import labels and make predictions

$report = new ConfusionMatrix();

$result = $report->generate($predictions, $labels);

var_dump($result);
array(3) {
  ["dog"]=> array(3) {
    ["dog"]=> int(842)
    ["cat"]=> int(5)
    ["turtle"]=> int(0)
  }
  ["cat"]=> array(3) {
    ["dog"]=> int(0)
    ["cat"]=> int(783)
    ["turtle"]=> int(3)
  }
  ["turtle"]=> array(2) {
    ["dog"]=> int(31)
    ["cat"]=> int(79)
    ["turtle"]=> int(496)
  }
}