Home  · Classes  · Annotated Classes  · Modules  · Members  · Namespaces  · Related Pages

SVMWrapper Class Reference

#include <OpenMS/ANALYSIS/SVM/SVMWrapper.h>

List of all members.


Detailed Description

Serves as a wrapper for the libsvm.

This class can be used for svm predictions. You can either perform classification or regression and choose certain kernel fuctions and additional parameters. Furthermore the models can be saved and loaded and we support also a new kernel function that was specially designed for learning with small sequences of different lengths.

Public Member Functions

 SVMWrapper ()
 standard constructor
 ~SVMWrapper ()
 destructor
void setParameter (SVM_parameter_type type, Int value)
 You can set the parameters of the svm:.
void setParameter (SVM_parameter_type type, DoubleReal value)
 sets the double parameters of the svm
Int train (struct svm_problem *problem)
 trains the svm
void saveModel (std::string modelFilename) const throw (Exception::UnableToCreateFile)
 saves the svm model
void loadModel (std::string modelFilename)
 loads the model
void predict (struct svm_problem *predictProblem, std::vector< DoubleReal > &predicted_rts)
 predicts the labels using the trained model
Int getIntParameter (SVM_parameter_type type)
 You can get the actual int- parameters of the svm.
DoubleReal getDoubleParameter (SVM_parameter_type type)
 You can get the actual double- parameters of the svm.
void predict (const std::vector< svm_node * > &vectors, std::vector< DoubleReal > &predicted_rts)
 predicts the labels using the trained model
DoubleReal performCrossValidation (svm_problem *problem, const std::map< SVM_parameter_type, DoubleReal > &start_values, const std::map< SVM_parameter_type, DoubleReal > &step_sizes, const std::map< SVM_parameter_type, DoubleReal > &end_values, UInt number_of_partitions, UInt number_of_runs, std::map< SVM_parameter_type, DoubleReal > &best_parameters, bool additive_step_size=true, bool output=false, String performances_file_name="performances.txt", bool mcc_as_performance_measure=false)
 Performs a CV for the data given by 'problem'.
DoubleReal getSVRProbability ()
 Returns the probability parameter sigma of the fitted laplace model.
void getSignificanceBorders (svm_problem *data, std::pair< DoubleReal, DoubleReal > &borders, DoubleReal confidence=0.95, UInt number_of_runs=10, UInt number_of_partitions=5, DoubleReal step_size=0.01, UInt max_iterations=1000000)
 calculates the significance borders of the error model and stores them in 'borders'
DoubleReal getPValue (DoubleReal sigma1, DoubleReal sigma2, std::pair< DoubleReal, DoubleReal > point)
 calculates a p-value for a given data point using the model parameters
void getDecisionValues (svm_problem *data, std::vector< DoubleReal > &decision_values)
 stores the prediction values for the encoded data in 'decision_values'
void scaleData (svm_problem *data, Int max_scale_value=-1)
 Scales the data such that every coloumn is scaled to [-1, 1].
svm_problem * computeKernelMatrix (svm_problem *problem1, svm_problem *problem2)
 computes the kernel matrix using the actual svm parameters and the given data
void setTrainingSample (svm_problem *training_sample)
 This is used for being able to perform predictions with non libsvm standard kernels.
void getSVCProbabilities (struct svm_problem *problem, std::vector< DoubleReal > &probabilities, std::vector< DoubleReal > &prediction_labels)
 This function fills probabilities with the probability estimates for the first class.
void setWeights (const std::vector< Int > &weight_labels, const std::vector< DoubleReal > &weights)
 Sets weights for the classes in C_SVC (see libsvm documentation for further details).

Static Public Member Functions

static void createRandomPartitions (svm_problem *problem, UInt number, std::vector< svm_problem * > &partitions)
 You can create 'number' equally sized random partitions.
static svm_problem * mergePartitions (const std::vector< svm_problem * > &problems, UInt except)
 You can merge partitions excuding the partition with index 'except'.
static void getLabels (svm_problem *problem, std::vector< DoubleReal > &labels)
 Stores the stored labels of the encoded SVM data at 'labels'.
static DoubleReal kernelOligo (const svm_node *x, const svm_node *y, const std::vector< DoubleReal > &gauss_table, DoubleReal sigma_square=0, UInt max_distance=50)
 calculates the oligo kernel value for the encoded sequences 'x' and 'y'
static void calculateGaussTable (UInt border_length, DoubleReal sigma, std::vector< DoubleReal > &gauss_table)

Protected Member Functions

void destroyProblem_ (svm_problem *problem)

Private Member Functions

UInt getNumberOfEnclosedPoints_ (DoubleReal m1, DoubleReal m2, const std::vector< std::pair< DoubleReal, DoubleReal > > &points)
void initParameters_ ()
 Initializes the svm with standard parameters.

Private Attributes

svm_parameter * param_
svm_model * model_
DoubleReal sigma_
std::vector< DoubleRealsigmas_
std::vector< DoubleRealgauss_table_
std::vector< std::vector
< DoubleReal > > 
gauss_tables_
UInt kernel_type_
UInt border_length_
svm_problem * training_set_
svm_problem * training_problem_


Constructor & Destructor Documentation

SVMWrapper (  ) 

standard constructor

~SVMWrapper (  ) 

destructor


Member Function Documentation

void setParameter ( SVM_parameter_type  type,
Int  value 
)

You can set the parameters of the svm:.

KERNEL_TYPE: can be LINEAR for the linear kernel RBF for the rbf kernel POLY for the polynomial kernel SIGMOID for the sigmoid kernel DEGREE: the degree for the polynomial- kernel and the locality- improved kernel

C: the C parameter of the svm

void setParameter ( SVM_parameter_type  type,
DoubleReal  value 
)

sets the double parameters of the svm

Int train ( struct svm_problem *  problem  ) 

trains the svm

The svm is trained with the data stored in the 'svm_problem' structure.

void saveModel ( std::string  modelFilename  )  const throw (Exception::UnableToCreateFile)

saves the svm model

The model of the trained svm is saved into 'modelFilename'. Throws an exception if the model cannot be saved.

void loadModel ( std::string  modelFilename  ) 

loads the model

The svm- model is loaded. After this, the svm is ready for prediction.

void predict ( struct svm_problem *  predictProblem,
std::vector< DoubleReal > &  predicted_rts 
)

predicts the labels using the trained model

The prediction process is started and the results are stored in 'predicted_rts'.

Int getIntParameter ( SVM_parameter_type  type  ) 

You can get the actual int- parameters of the svm.

KERNEL_TYPE: can be LINEAR for the linear kernel RBF for the rbf kernel POLY for the polynomial kernel SIGMOID for the sigmoid kernel

DEGREE: the degree for the polynomial- kernel and the locality- improved kernel

SVM_TYPE: the SVm type of the svm: can be NU_SVR or EPSILON_SVR

DoubleReal getDoubleParameter ( SVM_parameter_type  type  ) 

You can get the actual double- parameters of the svm.

C: the C parameter of the svm P: the P parameter of the svm (sets the epsilon in epsilon-svr) NU: the nu parameter in nu-SVR GAMMA: for POLY, RBF and SIGMOID

static void createRandomPartitions ( svm_problem *  problem,
UInt  number,
std::vector< svm_problem * > &  partitions 
) [static]

You can create 'number' equally sized random partitions.

This function creates 'number' equally sized random partitions and stores them in 'partitions'.

static svm_problem* mergePartitions ( const std::vector< svm_problem * > &  problems,
UInt  except 
) [static]

You can merge partitions excuding the partition with index 'except'.

void predict ( const std::vector< svm_node * > &  vectors,
std::vector< DoubleReal > &  predicted_rts 
)

predicts the labels using the trained model

The prediction process is started and the results are stored in 'predicted_rts'.

static void getLabels ( svm_problem *  problem,
std::vector< DoubleReal > &  labels 
) [static]

Stores the stored labels of the encoded SVM data at 'labels'.

DoubleReal performCrossValidation ( svm_problem *  problem,
const std::map< SVM_parameter_type, DoubleReal > &  start_values,
const std::map< SVM_parameter_type, DoubleReal > &  step_sizes,
const std::map< SVM_parameter_type, DoubleReal > &  end_values,
UInt  number_of_partitions,
UInt  number_of_runs,
std::map< SVM_parameter_type, DoubleReal > &  best_parameters,
bool  additive_step_size = true,
bool  output = false,
String  performances_file_name = "performances.txt",
bool  mcc_as_performance_measure = false 
)

Performs a CV for the data given by 'problem'.

DoubleReal getSVRProbability (  ) 

Returns the probability parameter sigma of the fitted laplace model.

The libsvm is used to fit a laplace model to the prediction values by performing an internal cv using the training set if setParameter(PROBABILITY, 1) was invoked before using train. Look for your libsvm documentation for more details. The model parameter sigma is returned by this method. If no model was fitted during training zero is returned.

static DoubleReal kernelOligo ( const svm_node *  x,
const svm_node *  y,
const std::vector< DoubleReal > &  gauss_table,
DoubleReal  sigma_square = 0,
UInt  max_distance = 50 
) [static]

calculates the oligo kernel value for the encoded sequences 'x' and 'y'

This kernel function calculates the oligo kernel value [Meinicke 04] for the sequences 'x' and 'y' that had been encoded by the encodeOligoBorder... function of the LibSVMEncoder class.

void getSignificanceBorders ( svm_problem *  data,
std::pair< DoubleReal, DoubleReal > &  borders,
DoubleReal  confidence = 0.95,
UInt  number_of_runs = 10,
UInt  number_of_partitions = 5,
DoubleReal  step_size = 0.01,
UInt  max_iterations = 1000000 
)

calculates the significance borders of the error model and stores them in 'borders'

DoubleReal getPValue ( DoubleReal  sigma1,
DoubleReal  sigma2,
std::pair< DoubleReal, DoubleReal point 
)

calculates a p-value for a given data point using the model parameters

Uses the model parameters to calculate the p-value for 'point' which has the data entries: measured, predicted retention time.

void getDecisionValues ( svm_problem *  data,
std::vector< DoubleReal > &  decision_values 
)

stores the prediction values for the encoded data in 'decision_values'

This function can be used to get the prediction values of the data if a model is already trained by the train() method. For regression the result is the same as for the method predict. For classification this function returns the distance from the separating hyperplane. For multiclass classification the decision_values vector will be empty.

void scaleData ( svm_problem *  data,
Int  max_scale_value = -1 
)

Scales the data such that every coloumn is scaled to [-1, 1].

Scales the x[][].value values of the svm_problem* structure. If the second parameter is omitted, the data is scaled to [-1, 1]. Otherwise the data is scaled to [0, max_scale_value]

static void calculateGaussTable ( UInt  border_length,
DoubleReal  sigma,
std::vector< DoubleReal > &  gauss_table 
) [static]

svm_problem* computeKernelMatrix ( svm_problem *  problem1,
svm_problem *  problem2 
)

computes the kernel matrix using the actual svm parameters and the given data

This function can be used to compute a kernel matrix. 'problem1' and 'problem2' are used together wit the oligo kernel function (could be extended if you want to use your own kernel functions).

void setTrainingSample ( svm_problem *  training_sample  ) 

This is used for being able to perform predictions with non libsvm standard kernels.

void getSVCProbabilities ( struct svm_problem *  problem,
std::vector< DoubleReal > &  probabilities,
std::vector< DoubleReal > &  prediction_labels 
)

This function fills probabilities with the probability estimates for the first class.

The libSVM function svm_predict_probability is called to get probability estimates for the positive class. Since this is only used for binary classification it is sufficient for every test example to report the probability of the test example belonging to the positive class. Probability estimates have to be turned on during training (svm.setParameter(PROBABILITY, 1)), otherwise this method will fill the 'probabilities' vector with -1s.

void setWeights ( const std::vector< Int > &  weight_labels,
const std::vector< DoubleReal > &  weights 
)

Sets weights for the classes in C_SVC (see libsvm documentation for further details).

void destroyProblem_ ( svm_problem *  problem  )  [protected]

UInt getNumberOfEnclosedPoints_ ( DoubleReal  m1,
DoubleReal  m2,
const std::vector< std::pair< DoubleReal, DoubleReal > > &  points 
) [private]

void initParameters_ (  )  [private]

Initializes the svm with standard parameters.


Member Data Documentation

svm_parameter* param_ [private]

svm_model* model_ [private]

DoubleReal sigma_ [private]

std::vector<DoubleReal> sigmas_ [private]

std::vector<DoubleReal> gauss_table_ [private]

std::vector<std::vector<DoubleReal> > gauss_tables_ [private]

UInt kernel_type_ [private]

UInt border_length_ [private]

svm_problem* training_set_ [private]

svm_problem* training_problem_ [private]


The documentation for this class was generated from the following file:
Generated Tue Apr 1 15:36:42 2008 -- using doxygen 1.5.4 OpenMS / TOPP 1.1