Detailed Description

Simple conjugate gradients optimizer.

Algorithm is as per Shanno and Phua, ACM Transactions On Mathematical Software 6 (December 1980), 618-622

Conjugate gradients optimization is sensitive to the scales of the derivatives of the various attributes being optimized. By default, the scales are estimated from the range of values found for the attribute upon initialization. These estimates can be viewed either by calling Model::get_range(my_float_key) or by turning on TERSE logging and looking at logged messages. If this estimate does not accurately reflect the scale, then you can use Model::set_range to set a more accurate range for the parameters.

Note: Currently, rigid bodies are not handled (and will not be moved by this optimizer).

Definition at line 34 of file core/ConjugateGradients.h.

Public Member Functions
	ConjugateGradients (Model *m, std::string name="ConjugateGradients%1%")

virtual Float	do_optimize (unsigned int max_steps) override
	override this function to do actual optimization More...

virtual std::string	get_type_name () const override

virtual ::IMP::VersionInfo	get_version_info () const override
	Get information about the module and version of the object. More...

void	set_gradient_threshold (Float t)
	Set the threshold for the minimum gradient. More...

void	set_max_change (Float t)
	Limit how far anything can change each time step. More...

Public Member Functions inherited from IMP::AttributeOptimizer
	AttributeOptimizer (Model *m, std::string name="Optimizer %1%")

Public Member Functions inherited from IMP::Optimizer
	Optimizer (Model *m, std::string name="Optimizer %1%")

double	get_last_score () const
	Return the score found in the last evaluate. More...

ScoringFunction *	get_scoring_function () const
	Return the scoring function that is being used. More...

bool	get_stop_on_good_score () const

double	optimize (unsigned int max_steps)
	Optimize the model for up to max_steps iterations. More...

virtual void	set_scoring_function (ScoringFunctionAdaptor sf)
	Set the scoring function to use. More...

void	set_stop_on_good_score (bool tf)

void	remove_optimizer_state (OptimizerState *d)

void	remove_optimizer_states (const OptimizerStates &d)

void	set_optimizer_states (const OptimizerStates &ps)

void	set_optimizer_states_order (const OptimizerStates &objs)

unsigned int	add_optimizer_state (OptimizerState *obj)

void	add_optimizer_states (const OptimizerStates &objs)

void	clear_optimizer_states ()

unsigned int	get_number_of_optimizer_states () const

bool	get_has_optimizer_states ()

OptimizerState *	get_optimizer_state (unsigned int i) const

OptimizerStates	get_optimizer_states () const

void	erase_optimizer_state (unsigned int i)

void	reserve_optimizer_states (unsigned int sz)

Public Member Functions inherited from IMP::ModelObject
	ModelObject (Model *m, std::string name)

bool	get_has_dependencies () const
	Return whether this object has dependencies computed. More...

bool	get_has_required_score_states () const
	Return whether score states are computed. More...

ModelObjectsTemp	get_inputs () const

ModelObjectsTemps	get_interactions () const
	Get the interacting sets induced by this ModelObject. More...

Model *	get_model () const

ModelObjectsTemp	get_outputs () const

const ScoreStatesTemp &	get_required_score_states () const
	Get the score states that are ancestors of this in the dependency graph. More...

void	set_has_dependencies (bool tf)
	Either invalidate the dependencies or ensure they are correct. More...

void	set_has_required_score_states (bool tf)
	Compute the required score states. More...

Public Member Functions inherited from IMP::Object
virtual void	clear_caches ()

CheckLevel	get_check_level () const

LogLevel	get_log_level () const

void	set_check_level (CheckLevel l)

void	set_log_level (LogLevel l)
	Set the logging level used in this object. More...

void	set_was_used (bool tf) const

void	show (std::ostream &out=std::cout) const

const std::string &	get_name () const

void	set_name (std::string name)

Additional Inherited Members
Public Attributes inherited from IMP::Optimizer
list	optimizer_states
	A Python list of OptimizerStates More...

Protected Member Functions inherited from IMP::AttributeOptimizer
double	get_width (FloatKey k) const

FloatIndexes	get_optimized_attributes () const

void	set_value (FloatIndex fi, double v) const

Float	get_value (FloatIndex fi) const

Float	get_derivative (FloatIndex fi) const

void	set_scaled_value (FloatIndex fi, Float v) const

double	get_scaled_value (FloatIndex fi) const

double	get_scaled_derivative (FloatIndex fi) const

void	clear_range_cache ()
	Clear the cache of range information. Do this at the start of. More...

Protected Member Functions inherited from IMP::Optimizer
virtual ModelObjectsTemp	do_get_inputs () const override

virtual ModelObjectsTemp	do_get_outputs () const override
	don't return anything here to avoid pointless dependencies More...

ModelObjectsTemp	get_optimizer_state_inputs () const

void	update_states () const
	Update optimizer states, should be called at each successful step. More...

Protected Member Functions inherited from IMP::ModelObject
virtual ModelObjectsTemps	do_get_interactions () const

virtual void	handle_set_has_required_score_states (bool)

Protected Member Functions inherited from IMP::Object
	Object (std::string name)
	Construct an object with the given name. More...

virtual void	do_destroy ()

Member Function Documentation

virtual Float IMP::core::ConjugateGradients::do_optimize ( unsigned int ns )

overridevirtual

override this function to do actual optimization

Implements IMP::Optimizer.

virtual ::IMP::VersionInfo IMP::core::ConjugateGradients::get_version_info ( ) const

overridevirtual

Get information about the module and version of the object.

Reimplemented from IMP::Object.

Definition at line 50 of file core/ConjugateGradients.h.

void IMP::core::ConjugateGradients::set_gradient_threshold ( Float t )

Set the threshold for the minimum gradient.

Definition at line 40 of file core/ConjugateGradients.h.

void IMP::core::ConjugateGradients::set_max_change ( Float t )

Limit how far anything can change each time step.

Definition at line 47 of file core/ConjugateGradients.h.

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

core/ConjugateGradients.h

Detailed Description

Public Member Functions

Additional Inherited Members

Member Function Documentation