loss.h File Reference

Declares the unified Loss class plus its Error and Regularization enumerations. More...

#include "neural_network.h"
#include "back_propagation.h"

Go to the source code of this file.

Classes
class	opennn::Loss
	Trainable loss function attached to a NeuralNetwork and a Dataset. More...
struct	opennn::Loss::EvaluationResult
	Output of calculate_error(). More...

Namespaces
namespace	opennn

Detailed Description

Declares the unified Loss class plus its Error and Regularization enumerations.

Loss couples a NeuralNetwork to a Dataset and implements forward evaluation, gradient back-propagation and regularization for every built-in loss function (MeanSquaredError, NormalizedSquaredError, WeightedSquaredError, CrossEntropy, CrossEntropy3d, MinkowskiError).