BALL  1.4.1
 All Classes Namespaces Files Functions Variables Typedefs Enumerations Enumerator Properties Friends Macros Groups Pages
List of all members
BALL::QSAR::PCRModel Class Reference

#include <BALL/QSAR/pcrModel.h>

Inheritance diagram for BALL::QSAR::PCRModel:
BALL::QSAR::LinearModel BALL::QSAR::LatentVariableModel BALL::QSAR::RegressionModel BALL::QSAR::Model

Public Member Functions

Constructors and Destructors
 PCRModel (const QSARData &q, double frac_var=0.99)
 ~PCRModel ()
- Public Member Functions inherited from BALL::QSAR::LinearModel
 LinearModel (const QSARData &q)
 ~LinearModel ()
virtual void operator= (const LinearModel &m)
virtual Vector< doublepredict (const vector< double > &substance, bool transform=1)
- Public Member Functions inherited from BALL::QSAR::RegressionModel
 RegressionModel (const QSARData &q)
 ~RegressionModel ()
virtual void operator= (const RegressionModel &m)
const BALL::Matrix< double > * getTrainingResult () const
virtual void saveToFile (string filename)
virtual void readFromFile (string filename)
void show ()
- Public Member Functions inherited from BALL::QSAR::Model
 Model (const QSARData &q)
virtual ~Model ()
virtual void operator= (const Model &m)
void copyData (const Model &m)
void copyDescriptorIDs (const Model &m)
void readTrainingData ()
void deleteDescriptorIDs ()
virtual bool optimizeParameters (int, int)
bool optimizeParameters (int k)
virtual double calculateStdErr ()
std::multiset< unsigned int > * getDescriptorIDs ()
void setDataSource (const QSARData *q)
const Matrix< double > * getDescriptorMatrix ()
const vector< string > * getSubstanceNames ()
const vector< string > * getDescriptorNames ()
const Matrix< double > * getY ()
void setDescriptorIDs (const std::multiset< unsigned int > &sl)
const string * getType ()
void getUnnormalizedFeatureValue (int compound, int feature, double &return_value)
void getUnnormalizedResponseValue (int compound, int response, double &return_value)
- Public Member Functions inherited from BALL::QSAR::LatentVariableModel
 LatentVariableModel ()
const Matrix< double > * getLatentVariables ()
const Matrix< double > * getWeights ()
const Matrix< double > * getLoadings ()

Protected Attributes

Attributes
double frac_var_
- Protected Attributes inherited from BALL::QSAR::LatentVariableModel
Matrix< doublelatent_variables_
Matrix< doubleloadings_
Matrix< doubleweights_

Accessors

void train ()
void setFracVar (double frac_var)
void setParameters (vector< double > &v)
vector< doublegetParameters () const
static void calculateEigenvectors (const Matrix< double > &data, double frac_var, Matrix< double > &output)

Additional Inherited Members

- Public Attributes inherited from BALL::QSAR::RegressionModel
RegressionValidationvalidation
- Public Attributes inherited from BALL::QSAR::Model
const QSARDatadata
Validationmodel_val
- Protected Member Functions inherited from BALL::QSAR::LinearModel
void calculateOffsets ()
- Protected Member Functions inherited from BALL::QSAR::RegressionModel
void readDescriptorInformationFromFile (std::ifstream &in, int no_descriptors, bool transformation, int no_coefficients)
void saveDescriptorInformationToFile (std::ofstream &out)

Detailed Description

Definition at line 46 of file pcrModel.h.

Constructor & Destructor Documentation

BALL::QSAR::PCRModel::PCRModel ( const QSARData q,
double  frac_var = 0.99 
)
constructur
Parameters
qQSAR-wrapper object, from which the data for this model should be taken
frac_varthe part of the variance that is to be explained by the latent variables; as many latent variables as necessary to achieve this are created
BALL::QSAR::PCRModel::~PCRModel ( )

Member Function Documentation

static void BALL::QSAR::PCRModel::calculateEigenvectors ( const Matrix< double > &  data,
double  frac_var,
Matrix< double > &  output 
)
static
calculates the first eigenvectors of the given matrix 'data' and saves them as columns of matrix 'output' 
Parameters
frac_varthe fraction of the variance that is to the covered the selected eigenvectors
vector<double> BALL::QSAR::PCRModel::getParameters ( ) const
virtual

Reimplemented from BALL::QSAR::Model.

void BALL::QSAR::PCRModel::setFracVar ( double  frac_var)

set the fraction of the variance that should be explained by the latente variables.

void BALL::QSAR::PCRModel::setParameters ( vector< double > &  )
virtual

sets the model parameters according to the given values.

Reimplemented from BALL::QSAR::Model.

void BALL::QSAR::PCRModel::train ( )
virtual

Starts principal component regression with the current data and saves the resulting linear combination of descriptors to training_result.

Implements BALL::QSAR::Model.

Member Data Documentation

double BALL::QSAR::PCRModel::frac_var_
protected

fraction of the variance that is to be explained

Definition at line 88 of file pcrModel.h.