magpie.models.regression.crystal

Class StructureKRRBasedRegression

java.lang.Object

magpie.models.BaseModel

magpie.models.regression.BaseRegression

magpie.models.regression.crystal.StructureKRRBasedRegression

All Implemented Interfaces:

java.io.Serializable, java.lang.Cloneable, AbstractRegressionModel, Citable, Commandable, Options, Printable, Savable

Direct Known Subclasses:

CoulombEwaldMatrixRegression, CoulombSineMatrixRegression, PRDFRegression

public abstract class StructureKRRBasedRegression
extends BaseRegression

Abstract class for methods that predict properties of crystal structures with KKR-based schemes. Data must be a CrystalStructureDataset.

This implementation currently uses CholeskyDecomposition to perform to solve the ridge regression.

How to Extend this Class

1. Implement computeRepresentation(vassal.data.Cell) to create a representation of a structure.
2. Implement computeSimiliarity(Object, Object) to compute the similarity between two structures based on their representations. Note: This operation should include your kernel function.
3. Implement Options.setOptions(java.util.List) and
4. Blame Logan Ward if there's something else he forgot to list!

Implemented Commands

match $<dataset> <n>: Find the entries in the training set that are closest to those in a user-provided dataset
dataset: Dataset to be matched
n: Number to print
Will print out the n entries in the dataset that are closest to the entries in the dataset.

See Also:

Serialized Form

Field Summary

Fields inherited from class magpie.models.BaseModel

AttributeSelector, trained, TrainingStats, validated, ValidationStats

Constructor Summary

Constructors

Constructor and Description
StructureKRRBasedRegression()

Method Summary

Modifier and Type	Method and Description
StructureKRRBasedRegression	clone()
abstract java.lang.Object	computeRepresentation(vassal.data.Cell strc) Given a structure, compute representation of the structure.
protected abstract double	computeSimiliarity(java.lang.Object strc1, java.lang.Object strc2) Compute similarity between two crystal structures.
java.util.List<java.lang.String>	findClosestEntries(CrystalStructureEntry entry, int num) Find the name of the training entries that are closest in similarity to this entry.
int	getNFittingParameters() Number of fitting parameters in a model.
protected java.lang.String	printModel_protected() Internal method that handles printing the model as a string.
void	run_protected(Dataset TrainData) Run a model without checking if stuff is trained (use carefully)
java.lang.Object	runCommand(java.util.List<java.lang.Object> Command) Process some command described by a list of Objects.
void	setLambda(double lambda) Set the ridge regularization parameter.
protected void	train_protected(Dataset TrainData) Train a model without evaluating performance

Methods inherited from class magpie.models.regression.BaseRegression

doRobustRegression, getRobustRegressionQ, robustTraining, setRobustRegressionQ

Methods inherited from class magpie.models.BaseModel

about, crossValidate, crossValidate, crossValidate, done, externallyValidate, getAttributeSelector, getCitations, getFilter, getTrainTime, getValidationMethod, handleSetCommand, isTrained, isValidated, loadState, printCommand, printDescription, printModel, printModelDescriptionDetails, resetModel, run, saveCommand, saveState, setAttributeSelector, setComponent, setFilter, train, train

Methods inherited from class java.lang.Object

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

Methods inherited from interface magpie.utility.interfaces.Options

printUsage, setOptions

Constructor Detail

StructureKRRBasedRegression

public StructureKRRBasedRegression()

Method Detail

clone

public StructureKRRBasedRegression clone()

Overrides:

clone in class BaseRegression

setLambda

public void setLambda(double lambda)

Set the ridge regularization parameter.

Parameters:

lambda -

train_protected

protected void train_protected(Dataset TrainData)

Description copied from class: BaseModel

Train a model without evaluating performance

Specified by:

train_protected in class BaseModel

Parameters:

TrainData - Training data

run_protected

public void run_protected(Dataset TrainData)

Description copied from class: BaseModel

Run a model without checking if stuff is trained (use carefully)

Specified by:

run_protected in class BaseModel

Parameters:

TrainData - Training data

computeSimiliarity

protected abstract double computeSimiliarity(java.lang.Object strc1,
                                             java.lang.Object strc2)

Compute similarity between two crystal structures.

Parameters:

strc1 - Representation of structure #1

strc2 - Representation of structure #2

Returns:

Similarity between two structures

computeRepresentation

public abstract java.lang.Object computeRepresentation(vassal.data.Cell strc)

Given a structure, compute representation of the structure. For instance, this code could be used to compute the RDF or Coulomb matrix.

Parameters:

strc - Structure

Returns:

Representation of the structure

findClosestEntries

public java.util.List<java.lang.String> findClosestEntries(CrystalStructureEntry entry,
                                                           int num)

Find the name of the training entries that are closest in similarity to this entry.

Parameters:

entry - Entry to be matched

num - Number to list

Returns:

List of names of closest training entries

printModel_protected

protected java.lang.String printModel_protected()

Description copied from class: BaseModel

Internal method that handles printing the model as a string. Should end with a "\n"

Specified by:

printModel_protected in class BaseModel

Returns:

String representation of model

getNFittingParameters

public int getNFittingParameters()

Description copied from interface: AbstractRegressionModel

Number of fitting parameters in a model. Only pertinent to nonlinear regression. Otherwise, this should just return 0.

Returns:

Number of fitting parameters

runCommand

public java.lang.Object runCommand(java.util.List<java.lang.Object> Command)
                            throws java.lang.Exception

Description copied from interface: Commandable

Process some command described by a list of Objects.

Specified by:

runCommand in interface Commandable

Overrides:

runCommand in class BaseRegression

Parameters:

Command - Command as a list of objects

Returns:

Any output (null if no output generated by command)

Throws:

java.lang.Exception - If something goes wrong