FirstOrderStochasticCostFunction类 参考

详细描述

The first order stochastic cost function base class.

The class gives the interface used in first order stochastic minimizers

The cost function must be Written as a finite sample-specific sum of cost. For example, least squares cost function,

\[ f(w)=\frac{ \sum_i{ (y_i-w^T x_i)^2 } }{2} \]

where \((y_i,x_i)\) is the i-th sample, \(y_i\) is the label and \(x_i\) is the features

在文件 FirstOrderStochasticCostFunction.h 第 50 行定义.

类 FirstOrderStochasticCostFunction 继承关系图:

Public 成员函数
virtual void	begin_sample ()=0
virtual bool	next_sample ()=0
virtual SGVector< float64_t >	get_gradient ()=0
virtual float64_t	get_cost ()=0
virtual SGVector< float64_t >	obtain_variable_reference ()=0

成员函数说明

virtual void begin_sample ( )

pure virtual

Initialize to generate a sample sequence

virtual float64_t get_cost ( )

pure virtual

Get the cost given current target variables

For least squares, that is the value of \(f(w)\).

返回

cost

实现了 FirstOrderCostFunction.

在 FirstOrderSAGCostFunction 内被实现.

virtual SGVector<float64_t> get_gradient ( )

pure virtual

Get the SAMPLE gradient value wrt target variables

WARNING This method does return \( \frac{\partial f_i(w) }{\partial w} \), instead of \(\sum_i{ \frac{\partial f_i(w) }{\partial w} }\)

For least squares cost function, that is the value of \(\frac{\partial f_i(w) }{\partial w}\) given \(w\) is known where the index \(i\) is obtained by next_sample()

返回

sample gradient of variables

实现了 FirstOrderCostFunction.

在 FirstOrderSAGCostFunction 内被实现.

virtual bool next_sample ( )

pure virtual

Get next sample

返回

false if reach the end of the sample sequence

virtual SGVector<float64_t> obtain_variable_reference ( )

pure virtualinherited

Obtain a reference of target variables Minimizers will modify target variables in place.

This method will be called by FirstOrderMinimizer::minimize()

For least squares, that is \(w\)

返回

reference of variables

---

该类的文档由以下文件生成:

• FirstOrderStochasticCostFunction.h