Sha256: cb41e3ea4542e8a6b56d58c24ca76af73ab7eaf159e0f0b0d31c04166fa5e051
Contents?: true
Size: 835 Bytes
Versions: 8
Compression:
Stored size: 835 Bytes
Contents
# frozen_string_literal: true
require 'rumale/base/evaluator'
module Rumale
module EvaluationMeasure
# MeanSquaredError is a class that calculates the mean squared error.
#
# @example
# require 'rumale/evaluation_measure/mean_squared_error'
#
# evaluator = Rumale::EvaluationMeasure::MeanSquaredError.new
# puts evaluator.score(ground_truth, predicted)
class MeanSquaredError
include ::Rumale::Base::Evaluator
# Calculate mean squared error.
#
# @param y_true [Numo::DFloat] (shape: [n_samples, n_outputs]) Ground truth target values.
# @param y_pred [Numo::DFloat] (shape: [n_samples, n_outputs]) Estimated target values.
# @return [Float] Mean squared error
def score(y_true, y_pred)
((y_true - y_pred)**2).mean
end
end
end
end
Version data entries
8 entries across 8 versions & 1 rubygems