Class: Rumale::EvaluationMeasure::Accuracy

Inherits:

Object

• Object
• Rumale::EvaluationMeasure::Accuracy

Includes:

Base::Evaluator

Defined in:

rumale-evaluation_measure/lib/rumale/evaluation_measure/accuracy.rb

Overview

Accuracy is a class that calculates the accuracy of classifier from the predicted labels.

Examples:

require 'rumale/evaluation_measure/accuracy'

evaluator = Rumale::EvaluationMeasure::Accuracy.new
puts evaluator.score(ground_truth, predicted)

Instance Method Summary

• #score(y_true, y_pred) ⇒ Float

  Calculate mean accuracy.

Instance Method Details

#score(y_true, y_pred) ⇒ Float

Calculate mean accuracy.

Parameters:

• y_true (Numo::Int32) —

  (shape: [n_samples]) Ground truth labels.

• y_pred (Numo::Int32) —

  (shape: [n_samples]) Predicted labels.

Returns:

• (Float) —

  Mean accuracy

# File 'rumale-evaluation_measure/lib/rumale/evaluation_measure/accuracy.rb', line 22

def score(y_true, y_pred)
  (y_true.to_a.map.with_index { |label, n| label == y_pred[n] ? 1 : 0 }).sum / y_true.size.to_f
end