OpenANN::Optimizer Class Reference

The common interface of all optimization algorithms. More...

#include <Optimizer.h>

Inheritance diagram for OpenANN::Optimizer:

Public Member Functions
virtual	~Optimizer ()
virtual void	setOptimizable (Optimizable &optimizable)=0
	Pass the objective function. More...
virtual void	setStopCriteria (const StoppingCriteria &sc)=0
	Pass the stop criteria. More...
virtual void	optimize ()=0
	Optimize until the optimization meets the stop criteria. More...
virtual bool	step ()=0
	Execute one optimization step. More...
virtual Eigen::VectorXd	result ()=0
	Determine the best result. More...
virtual std::string	name ()=0
	Get the name of the optimization algorithm. More...

Detailed Description

The common interface of all optimization algorithms.

Constructor & Destructor Documentation

virtual OpenANN::Optimizer::~Optimizer ( )

inlinevirtual

Member Function Documentation

virtual std::string OpenANN::Optimizer::name ( )

pure virtual

Get the name of the optimization algorithm.

Returns

name of the optimization algorithm

Implemented in OpenANN::MBSGD, OpenANN::IPOPCMAES, OpenANN::LMA, OpenANN::LBFGS, and OpenANN::CG.

virtual void OpenANN::Optimizer::optimize ( )

pure virtual

Optimize until the optimization meets the stop criteria.

Implemented in OpenANN::MBSGD, OpenANN::IPOPCMAES, OpenANN::LMA, OpenANN::LBFGS, and OpenANN::CG.

virtual Eigen::VectorXd OpenANN::Optimizer::result ( )

pure virtual

Determine the best result.

Returns

the best parameter the algorithm found

Implemented in OpenANN::MBSGD, OpenANN::IPOPCMAES, OpenANN::LMA, OpenANN::LBFGS, and OpenANN::CG.

virtual void OpenANN::Optimizer::setOptimizable ( Optimizable &  optimizable)

pure virtual

Pass the objective function.

Parameters

optimizable

objective function, e. g. error function of an ANN

Implemented in OpenANN::MBSGD, OpenANN::IPOPCMAES, OpenANN::LMA, OpenANN::LBFGS, and OpenANN::CG.

virtual void OpenANN::Optimizer::setStopCriteria ( const StoppingCriteria &  sc)

pure virtual

Pass the stop criteria.

Parameters

sc	the parameters used to stop the optimization

Implemented in OpenANN::MBSGD, OpenANN::IPOPCMAES, OpenANN::LMA, OpenANN::LBFGS, and OpenANN::CG.

virtual bool OpenANN::Optimizer::step ( )

pure virtual

Execute one optimization step.

Implemented in OpenANN::MBSGD, OpenANN::IPOPCMAES, OpenANN::LMA, OpenANN::LBFGS, and OpenANN::CG.

---

The documentation for this class was generated from the following file:

• OpenANN/optimization/Optimizer.h