NLOPTMinimizer类 参考

详细描述

The class wraps the external NLOPT library.

This minimizer supports bound constrainted minimization and unconstrainted minimization using the NLOPT library

在文件 NLOPTMinimizer.h 第 47 行定义.

类 NLOPTMinimizer 继承关系图:

Public 成员函数
	NLOPTMinimizer ()
	NLOPTMinimizer (FirstOrderCostFunction *fun)
virtual	~NLOPTMinimizer ()
virtual float64_t	minimize ()
virtual bool	supports_batch_update () const
virtual CMinimizerContext *	save_to_context ()
virtual void	load_from_context (CMinimizerContext *context)
virtual void	set_nlopt_parameters (nlopt_algorithm algorithm=NLOPT_LD_LBFGS, float64_t max_iterations=1000, float64_t variable_tolerance=1e-6, float64_t function_tolerance=1e-6)
virtual void	set_cost_function (FirstOrderCostFunction *fun)
virtual void	set_penalty_weight (float64_t penalty_weight)
virtual void	set_penalty_type (Penalty *penalty_type)

Protected 成员函数
virtual void	init_minimization ()
virtual void	update_context (CMinimizerContext *context)
virtual float64_t	get_penalty (SGVector< float64_t > var)
virtual void	update_gradient (SGVector< float64_t > gradient, SGVector< float64_t > var)

Protected 属性
SGVector< float64_t >	m_target_variable
float64_t	m_max_iterations
float64_t	m_variable_tolerance
float64_t	m_function_tolerance
nlopt_algorithm	m_nlopt_algorithm
FirstOrderCostFunction *	m_fun
Penalty *	m_penalty_type
float64_t	m_penalty_weight

构造及析构函数说明

NLOPTMinimizer

(

)

Default constructor

在文件 NLOPTMinimizer.cpp 第 37 行定义.

NLOPTMinimizer

(

FirstOrderCostFunction *

fun

)

Constructor

参数

fun	cost function

在文件 NLOPTMinimizer.cpp 第 47 行定义.

~NLOPTMinimizer ( )

virtual

Destructor

在文件 NLOPTMinimizer.cpp 第 43 行定义.

成员函数说明

virtual float64_t get_penalty ( SGVector< float64_t >  var )

protectedvirtualinherited

Get the penalty given target variables For L2 penalty, the target variable is \(w\) and the value of penalty is \(\lambda \frac{w^t w}{2}\), where \(\lambda\) is the weight of penalty

参数

var	the variable used in regularization

在文件 FirstOrderMinimizer.h 第 164 行定义.

void init_minimization ( )

protectedvirtual

在文件 NLOPTMinimizer.cpp 第 163 行定义.

virtual void load_from_context ( CMinimizerContext *  context )

virtual

Load the given context object to restores mutable variables Usually it is used in deserialization.

参数

context

a context object

重载 FirstOrderMinimizer .

在文件 NLOPTMinimizer.h 第 85 行定义.

float64_t minimize ( )

virtual

Do minimization and get the optimal value

返回

optimal value

实现了 FirstOrderMinimizer.

在文件 NLOPTMinimizer.cpp 第 65 行定义.

virtual CMinimizerContext* save_to_context ( )

virtual

Return a context object which stores mutable variables Usually it is used in serialization.

返回

a context object

重载 FirstOrderMinimizer .

在文件 NLOPTMinimizer.h 第 78 行定义.

virtual void set_cost_function ( FirstOrderCostFunction *  fun )

virtualinherited

Set cost function used in the minimizer

参数

fun	the cost function

在文件 FirstOrderMinimizer.h 第 92 行定义.

void set_nlopt_parameters ( nlopt_algorithm  algorithm = NLOPT_LD_LBFGS, float64_t  max_iterations = 1000, float64_t  variable_tolerance = 1e-6, float64_t  function_tolerance = 1e-6  )

virtual

在文件 NLOPTMinimizer.cpp 第 133 行定义.

virtual void set_penalty_type ( Penalty *  penalty_type )

virtualinherited

Set the type of penalty For example, L2 penalty

参数

penalty_type

the type of penalty. If NULL is given, regularization is not enabled.

在文件 FirstOrderMinimizer.h 第 137 行定义.

virtual void set_penalty_weight ( float64_t  penalty_weight )

virtualinherited

Set the weight of penalty

参数

penalty_weight

the weight of penalty, which is positive

在文件 FirstOrderMinimizer.h 第 126 行定义.

virtual bool supports_batch_update ( ) const

virtual

Does minimizer support batch update?

返回

whether minimizer supports batch update

实现了 FirstOrderMinimizer.

在文件 NLOPTMinimizer.h 第 71 行定义.

virtual void update_context ( CMinimizerContext *  context )

protectedvirtualinherited

Update a context object to store mutable variables

参数

context

a context object

被 FirstOrderStochasticMinimizer, SMDMinimizer , 以及 SMIDASMinimizer 重载.

在文件 FirstOrderMinimizer.h 第 147 行定义.

virtual void update_gradient ( SGVector< float64_t >  gradient, SGVector< float64_t >  var  )

protectedvirtualinherited

Add gradient of the penalty wrt target variables to unpenalized gradient For least sqaure with L2 penalty,

\[ L2f(w)=f(w) + L2(w) \]

where \( f(w)=\sum_i{(y_i-w^T x_i)^2}\) is the least sqaure cost function and \(L2(w)=\lambda \frac{w^t w}{2}\) is the L2 penalty

Target variables is \(w\) Unpenalized gradient is \(\frac{\partial f(w) }{\partial w}\) Gradient of the penalty wrt target variables is \(\frac{\partial L2(w) }{\partial w}\)

参数

gradient	unpenalized gradient wrt its target variable
var	the target variable

在文件 FirstOrderMinimizer.h 第 190 行定义.

类成员变量说明

FirstOrderCostFunction* m_fun

protectedinherited

Cost function

在文件 FirstOrderMinimizer.h 第 205 行定义.

float64_t m_function_tolerance

protected

absolute tolerance on function value

在文件 NLOPTMinimizer.h 第 123 行定义.

float64_t m_max_iterations

protected

max number of iterations

在文件 NLOPTMinimizer.h 第 117 行定义.

nlopt_algorithm m_nlopt_algorithm

protected

algorithm provided by NLOPT for minimization

在文件 NLOPTMinimizer.h 第 126 行定义.

Penalty* m_penalty_type

protectedinherited

the type of penalty

在文件 FirstOrderMinimizer.h 第 208 行定义.

float64_t m_penalty_weight

protectedinherited

the weight of penalty

在文件 FirstOrderMinimizer.h 第 211 行定义.

SGVector<float64_t> m_target_variable

protected

在文件 NLOPTMinimizer.h 第 111 行定义.

float64_t m_variable_tolerance

protected

absolute tolerance on optimization parameters

在文件 NLOPTMinimizer.h 第 120 行定义.

---

该类的文档由以下文件生成:

• NLOPTMinimizer.h
• NLOPTMinimizer.cpp