Package ffx.xray

Class ScaleBulkMinimize

java.lang.Object

ffx.xray.ScaleBulkMinimize

All Implemented Interfaces:

Terminatable, OptimizationListener

public class ScaleBulkMinimize
extends Object
implements OptimizationListener, Terminatable

ScaleBulkMinimize class.

Since:

1.0

Author:

Timothy D. Fenn

Constructor Summary

Constructors

Constructor	Description
ScaleBulkMinimize(ReflectionList reflectionList, DiffractionRefinementData refinementData, CrystalReciprocalSpace crystalReciprocalSpace, ParallelTeam parallelTeam)	Constructor for ScaleBulkMinimize.

Method Summary

Modifier and Type	Method	Description
double[]	getCoordinates(double[] x)	getCoordinates.
int	getNumberOfVariables()	Retrieves the number of variables used in the optimization process.
void	gridOptimizeBulkSolventModel()	Perform a grid optimization of the bulk solvent model to minimize the residual energy and improve refinement parameters.
void	gridOptimizeKsBs()	Perform a grid search optimization for bulk solvent and scaling parameters.
ScaleBulkEnergy	minimize(double eps)	Performs a minimization procedure with the specified convergence criterion (epsilon).
ScaleBulkEnergy	minimize(int m, double eps)	Performs a minimization process to optimize scaling and bulk solvent energy parameters.
boolean	optimizationUpdate(int iter, int nBFGS, int nfun, double grms, double xrms, double f, double df, double angle, LineSearch.LineSearchResult info)	This method is called by the optimizer after each step.
void	terminate()	This should be implemented as a blocking interrupt; when the method returns the Terminatable algorithm has reached a clean termination point.

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Details

ScaleBulkMinimize

public ScaleBulkMinimize(ReflectionList reflectionList,
 DiffractionRefinementData refinementData,
 CrystalReciprocalSpace crystalReciprocalSpace,
 ParallelTeam parallelTeam)

Constructor for ScaleBulkMinimize.

Parameters:

reflectionList - a ReflectionList object.

refinementData - a DiffractionRefinementData object.

crystalReciprocalSpace - a CrystalReciprocalSpace object.

parallelTeam - the ParallelTeam to execute the ScaleBulkMinimize.

Method Details

getCoordinates

public double[] getCoordinates(@Nullable
 double[] x)

getCoordinates.

Parameters:

x - the array to populate with parameters or null to create a new array.

Returns:

an array containing the parameters.

getNumberOfVariables

public int getNumberOfVariables()

Retrieves the number of variables used in the optimization process.

Returns:

the number of variables.

minimize

public ScaleBulkEnergy minimize(double eps)

Performs a minimization procedure with the specified convergence criterion (epsilon).

Parameters:

eps - the convergence threshold for the minimization process; smaller values indicate tighter convergence criteria.

Returns:

a ScaleBulkEnergy object resulting from the minimization process.

minimize

public ScaleBulkEnergy minimize(int m,
 double eps)

Performs a minimization process to optimize scaling and bulk solvent energy parameters.

Parameters:

m - the number of prior gradient evaluations used by the L-BFGS algorithm; larger values can improve convergence for challenging optimizations, but may increase memory usage.

eps - the convergence threshold for the minimization process; smaller values indicate tighter convergence criteria.

Returns:

the ScaleBulkEnergy object representing the refined bulk solvent energy after the minimization process.

optimizationUpdate

public boolean optimizationUpdate(int iter,
 int nBFGS,
 int nfun,
 double grms,
 double xrms,
 double f,
 double df,
 double angle,
 LineSearch.LineSearchResult info)

This method is called by the optimizer after each step.

It can be used to log status messages, update the user interface, or gracefully terminate the optimizer.

Specified by:

optimizationUpdate in interface OptimizationListener

Parameters:

iter - Number of iterations.

nBFGS - Number of L-BFGS correction vectors.

nfun - Number of function evaluations.

grms - RMS gradient at current solution.

xrms - RMS coordinate change at current solution.

f - Function value at current solution.

df - Change in the function value compared to the previous solution.

angle - Current angle between gradient and search direction.

info - Result of the line search (null at iteration == 0).

Returns:

A return value of false will terminate the optimization.

terminate

public void terminate()

This should be implemented as a blocking interrupt; when the method returns the Terminatable algorithm has reached a clean termination point. For example, between minimize or molecular dynamics steps.

Specified by:

terminate in interface Terminatable

gridOptimizeBulkSolventModel

public void gridOptimizeBulkSolventModel()

Perform a grid optimization of the bulk solvent model to minimize the residual energy and improve refinement parameters. This method adjusts the bulk solvent parameters (A and B) within a specified range using a grid search algorithm, evaluates the associated energy, and identifies optimal parameters based on the minimum residual.

gridOptimizeKsBs

public void gridOptimizeKsBs()

Perform a grid search optimization for bulk solvent and scaling parameters. The method adjusts the scaling factor and bulk solvent parameters (Ks and Bs) to minimize the energy function and improve the refinement results.