Gradient Ascent in 1 Dimension with Temperature Works for all functions that are continous with a specified derivative function Only finds the closest local max. More...

Public Member Functions
	GradientAscent1D (Gradient1DConfig *config)

bool	singleStep ()
	Runs a single step of the optimization algorithm. More...

void	execute ()
	Executes the full optimization algorithm. More...

float	getXValue ()

float	getValue ()

bool	continueExecution ()

Detailed Description

Gradient Ascent in 1 Dimension with Temperature Works for all functions that are continous with a specified derivative function Only finds the closest local max.

Temperature controls X movement around discontinuity in the derivative function EX: X=0 when F(x) = abs(x)

Use Example: GradientAscent1D ga(& [Gradient1DConfig]); ga.execute(); ga.getValue();

Member Function Documentation

◆ continueExecution()

bool GradientAscent1D::continueExecution ( )

Returns: Should continue execution?

◆ execute()

void GradientAscent1D::execute ( )

Executes the full optimization algorithm.

Note: This function is used when there is no need to step through each iteration. Most uses of GradientAscent1D will call this function

◆ getValue()

float GradientAscent1D::getValue ( )

Returns: the current guess of the max value

◆ getXValue()

float GradientAscent1D::getXValue ( )

Returns: the X value of the current guess of the max

◆ singleStep()

bool GradientAscent1D::singleStep ( )

Runs a single step of the optimization algorithm.

Note: This is most likely not the function you are looking for to find the max. This is used when executing multiple GradientAscent1D's in parallel

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

soccer/optimization/GradientAscent1D.hpp
soccer/optimization/GradientAscent1D.cpp