OpenCV  3.4.2
Open Source Computer Vision
Public Member Functions | Static Public Member Functions | List of all members
cv::ml::TrainData Class Referenceabstract

Class encapsulating training data. More...

#include "ml.hpp"

Public Member Functions

virtual ~TrainData ()
 
virtual int getCatCount (int vi) const =0
 
virtual Mat getCatMap () const =0
 
virtual Mat getCatOfs () const =0
 
virtual Mat getClassLabels () const =0
 Returns the vector of class labels. More...
 
virtual Mat getDefaultSubstValues () const =0
 
virtual int getLayout () const =0
 
virtual Mat getMissing () const =0
 
virtual int getNAllVars () const =0
 
void getNames (std::vector< String > &names) const
 Returns vector of symbolic names captured in loadFromCSV() More...
 
virtual Mat getNormCatResponses () const =0
 
virtual void getNormCatValues (int vi, InputArray sidx, int *values) const =0
 
virtual int getNSamples () const =0
 
virtual int getNTestSamples () const =0
 
virtual int getNTrainSamples () const =0
 
virtual int getNVars () const =0
 
virtual Mat getResponses () const =0
 
virtual int getResponseType () const =0
 
virtual void getSample (InputArray varIdx, int sidx, float *buf) const =0
 
virtual Mat getSamples () const =0
 
virtual Mat getSampleWeights () const =0
 
virtual Mat getTestNormCatResponses () const =0
 
virtual Mat getTestResponses () const =0
 
virtual Mat getTestSampleIdx () const =0
 
Mat getTestSamples () const
 Returns matrix of test samples. More...
 
virtual Mat getTestSampleWeights () const =0
 
virtual Mat getTrainNormCatResponses () const =0
 Returns the vector of normalized categorical responses. More...
 
virtual Mat getTrainResponses () const =0
 Returns the vector of responses. More...
 
virtual Mat getTrainSampleIdx () const =0
 
virtual Mat getTrainSamples (int layout=ROW_SAMPLE, bool compressSamples=true, bool compressVars=true) const =0
 Returns matrix of train samples. More...
 
virtual Mat getTrainSampleWeights () const =0
 
virtual void getValues (int vi, InputArray sidx, float *values) const =0
 
virtual Mat getVarIdx () const =0
 
Mat getVarSymbolFlags () const
 
virtual Mat getVarType () const =0
 
virtual void setTrainTestSplit (int count, bool shuffle=true)=0
 Splits the training data into the training and test parts. More...
 
virtual void setTrainTestSplitRatio (double ratio, bool shuffle=true)=0
 Splits the training data into the training and test parts. More...
 
virtual void shuffleTrainTest ()=0
 

Static Public Member Functions

static Ptr< TrainDatacreate (InputArray samples, int layout, InputArray responses, InputArray varIdx=noArray(), InputArray sampleIdx=noArray(), InputArray sampleWeights=noArray(), InputArray varType=noArray())
 Creates training data from in-memory arrays. More...
 
static Mat getSubVector (const Mat &vec, const Mat &idx)
 
static Ptr< TrainDataloadFromCSV (const String &filename, int headerLineCount, int responseStartIdx=-1, int responseEndIdx=-1, const String &varTypeSpec=String(), char delimiter=',', char missch='?')
 Reads the dataset from a .csv file and returns the ready-to-use training data. More...
 
static float missingValue ()
 

Detailed Description

Class encapsulating training data.

Please note that the class only specifies the interface of training data, but not implementation. All the statistical model classes in ml module accepts Ptr<TrainData> as parameter. In other words, you can create your own class derived from TrainData and pass smart pointer to the instance of this class into StatModel::train.

See also
Training Data

Constructor & Destructor Documentation

§ ~TrainData()

virtual cv::ml::TrainData::~TrainData ( )
virtual

Member Function Documentation

§ create()

static Ptr<TrainData> cv::ml::TrainData::create ( InputArray  samples,
int  layout,
InputArray  responses,
InputArray  varIdx = noArray(),
InputArray  sampleIdx = noArray(),
InputArray  sampleWeights = noArray(),
InputArray  varType = noArray() 
)
static
Python:
retval=cv.ml.TrainData_create(samples, layout, responses[, varIdx[, sampleIdx[, sampleWeights[, varType]]]])

Creates training data from in-memory arrays.

Parameters
samplesmatrix of samples. It should have CV_32F type.
layoutsee ml::SampleTypes.
responsesmatrix of responses. If the responses are scalar, they should be stored as a single row or as a single column. The matrix should have type CV_32F or CV_32S (in the former case the responses are considered as ordered by default; in the latter case - as categorical)
varIdxvector specifying which variables to use for training. It can be an integer vector (CV_32S) containing 0-based variable indices or byte vector (CV_8U) containing a mask of active variables.
sampleIdxvector specifying which samples to use for training. It can be an integer vector (CV_32S) containing 0-based sample indices or byte vector (CV_8U) containing a mask of training samples.
sampleWeightsoptional vector with weights for each sample. It should have CV_32F type.
varTypeoptional vector of type CV_8U and size <number_of_variables_in_samples> + <number_of_variables_in_responses>, containing types of each input and output variable. See ml::VariableTypes.

§ getCatCount()

virtual int cv::ml::TrainData::getCatCount ( int  vi) const
pure virtual
Python:
retval=cv.ml_TrainData.getCatCount(vi)

§ getCatMap()

virtual Mat cv::ml::TrainData::getCatMap ( ) const
pure virtual
Python:
retval=cv.ml_TrainData.getCatMap()

§ getCatOfs()

virtual Mat cv::ml::TrainData::getCatOfs ( ) const
pure virtual
Python:
retval=cv.ml_TrainData.getCatOfs()

§ getClassLabels()

virtual Mat cv::ml::TrainData::getClassLabels ( ) const
pure virtual
Python:
retval=cv.ml_TrainData.getClassLabels()

Returns the vector of class labels.

The function returns vector of unique labels occurred in the responses.

§ getDefaultSubstValues()

virtual Mat cv::ml::TrainData::getDefaultSubstValues ( ) const
pure virtual
Python:
retval=cv.ml_TrainData.getDefaultSubstValues()

§ getLayout()

virtual int cv::ml::TrainData::getLayout ( ) const
pure virtual
Python:
retval=cv.ml_TrainData.getLayout()

§ getMissing()

virtual Mat cv::ml::TrainData::getMissing ( ) const
pure virtual
Python:
retval=cv.ml_TrainData.getMissing()

§ getNAllVars()

virtual int cv::ml::TrainData::getNAllVars ( ) const
pure virtual
Python:
retval=cv.ml_TrainData.getNAllVars()

§ getNames()

void cv::ml::TrainData::getNames ( std::vector< String > &  names) const
Python:
None=cv.ml_TrainData.getNames(names)

Returns vector of symbolic names captured in loadFromCSV()

§ getNormCatResponses()

virtual Mat cv::ml::TrainData::getNormCatResponses ( ) const
pure virtual
Python:
retval=cv.ml_TrainData.getNormCatResponses()

§ getNormCatValues()

virtual void cv::ml::TrainData::getNormCatValues ( int  vi,
InputArray  sidx,
int *  values 
) const
pure virtual

§ getNSamples()

virtual int cv::ml::TrainData::getNSamples ( ) const
pure virtual
Python:
retval=cv.ml_TrainData.getNSamples()

§ getNTestSamples()

virtual int cv::ml::TrainData::getNTestSamples ( ) const
pure virtual
Python:
retval=cv.ml_TrainData.getNTestSamples()

§ getNTrainSamples()

virtual int cv::ml::TrainData::getNTrainSamples ( ) const
pure virtual
Python:
retval=cv.ml_TrainData.getNTrainSamples()

§ getNVars()

virtual int cv::ml::TrainData::getNVars ( ) const
pure virtual
Python:
retval=cv.ml_TrainData.getNVars()

§ getResponses()

virtual Mat cv::ml::TrainData::getResponses ( ) const
pure virtual
Python:
retval=cv.ml_TrainData.getResponses()

§ getResponseType()

virtual int cv::ml::TrainData::getResponseType ( ) const
pure virtual
Python:
retval=cv.ml_TrainData.getResponseType()

§ getSample()

virtual void cv::ml::TrainData::getSample ( InputArray  varIdx,
int  sidx,
float *  buf 
) const
pure virtual
Python:
None=cv.ml_TrainData.getSample(varIdx, sidx, buf)

§ getSamples()

virtual Mat cv::ml::TrainData::getSamples ( ) const
pure virtual
Python:
retval=cv.ml_TrainData.getSamples()

§ getSampleWeights()

virtual Mat cv::ml::TrainData::getSampleWeights ( ) const
pure virtual
Python:
retval=cv.ml_TrainData.getSampleWeights()

§ getSubVector()

static Mat cv::ml::TrainData::getSubVector ( const Mat vec,
const Mat idx 
)
static
Python:
retval=cv.ml.TrainData_getSubVector(vec, idx)

§ getTestNormCatResponses()

virtual Mat cv::ml::TrainData::getTestNormCatResponses ( ) const
pure virtual
Python:
retval=cv.ml_TrainData.getTestNormCatResponses()

§ getTestResponses()

virtual Mat cv::ml::TrainData::getTestResponses ( ) const
pure virtual
Python:
retval=cv.ml_TrainData.getTestResponses()

§ getTestSampleIdx()

virtual Mat cv::ml::TrainData::getTestSampleIdx ( ) const
pure virtual
Python:
retval=cv.ml_TrainData.getTestSampleIdx()

§ getTestSamples()

Mat cv::ml::TrainData::getTestSamples ( ) const
Python:
retval=cv.ml_TrainData.getTestSamples()

Returns matrix of test samples.

§ getTestSampleWeights()

virtual Mat cv::ml::TrainData::getTestSampleWeights ( ) const
pure virtual
Python:
retval=cv.ml_TrainData.getTestSampleWeights()

§ getTrainNormCatResponses()

virtual Mat cv::ml::TrainData::getTrainNormCatResponses ( ) const
pure virtual
Python:
retval=cv.ml_TrainData.getTrainNormCatResponses()

Returns the vector of normalized categorical responses.

The function returns vector of responses. Each response is integer from 0 to <number of classes>-1. The actual label value can be retrieved then from the class label vector, see TrainData::getClassLabels.

§ getTrainResponses()

virtual Mat cv::ml::TrainData::getTrainResponses ( ) const
pure virtual
Python:
retval=cv.ml_TrainData.getTrainResponses()

Returns the vector of responses.

The function returns ordered or the original categorical responses. Usually it's used in regression algorithms.

§ getTrainSampleIdx()

virtual Mat cv::ml::TrainData::getTrainSampleIdx ( ) const
pure virtual
Python:
retval=cv.ml_TrainData.getTrainSampleIdx()

§ getTrainSamples()

virtual Mat cv::ml::TrainData::getTrainSamples ( int  layout = ROW_SAMPLE,
bool  compressSamples = true,
bool  compressVars = true 
) const
pure virtual
Python:
retval=cv.ml_TrainData.getTrainSamples([, layout[, compressSamples[, compressVars]]])

Returns matrix of train samples.

Parameters
layoutThe requested layout. If it's different from the initial one, the matrix is transposed. See ml::SampleTypes.
compressSamplesif true, the function returns only the training samples (specified by sampleIdx)
compressVarsif true, the function returns the shorter training samples, containing only the active variables.

In current implementation the function tries to avoid physical data copying and returns the matrix stored inside TrainData (unless the transposition or compression is needed).

§ getTrainSampleWeights()

virtual Mat cv::ml::TrainData::getTrainSampleWeights ( ) const
pure virtual
Python:
retval=cv.ml_TrainData.getTrainSampleWeights()

§ getValues()

virtual void cv::ml::TrainData::getValues ( int  vi,
InputArray  sidx,
float *  values 
) const
pure virtual
Python:
None=cv.ml_TrainData.getValues(vi, sidx, values)

§ getVarIdx()

virtual Mat cv::ml::TrainData::getVarIdx ( ) const
pure virtual
Python:
retval=cv.ml_TrainData.getVarIdx()

§ getVarSymbolFlags()

Mat cv::ml::TrainData::getVarSymbolFlags ( ) const
Python:
retval=cv.ml_TrainData.getVarSymbolFlags()

§ getVarType()

virtual Mat cv::ml::TrainData::getVarType ( ) const
pure virtual
Python:
retval=cv.ml_TrainData.getVarType()

§ loadFromCSV()

static Ptr<TrainData> cv::ml::TrainData::loadFromCSV ( const String filename,
int  headerLineCount,
int  responseStartIdx = -1,
int  responseEndIdx = -1,
const String varTypeSpec = String(),
char  delimiter = ',',
char  missch = '?' 
)
static

Reads the dataset from a .csv file and returns the ready-to-use training data.

Parameters
filenameThe input file name
headerLineCountThe number of lines in the beginning to skip; besides the header, the function also skips empty lines and lines staring with #
responseStartIdxIndex of the first output variable. If -1, the function considers the last variable as the response
responseEndIdxIndex of the last output variable + 1. If -1, then there is single response variable at responseStartIdx.
varTypeSpecThe optional text string that specifies the variables' types. It has the format ord[n1-n2,n3,n4-n5,...]cat[n6,n7-n8,...]. That is, variables from n1 to n2 (inclusive range), n3, n4 to n5 ... are considered ordered and n6, n7 to n8 ... are considered as categorical. The range [n1..n2] + [n3] + [n4..n5] + ... + [n6] + [n7..n8] should cover all the variables. If varTypeSpec is not specified, then algorithm uses the following rules:
  • all input variables are considered ordered by default. If some column contains has non- numerical values, e.g. 'apple', 'pear', 'apple', 'apple', 'mango', the corresponding variable is considered categorical.
  • if there are several output variables, they are all considered as ordered. Error is reported when non-numerical values are used.
  • if there is a single output variable, then if its values are non-numerical or are all integers, then it's considered categorical. Otherwise, it's considered ordered.
delimiterThe character used to separate values in each line.
misschThe character used to specify missing measurements. It should not be a digit. Although it's a non-numerical value, it surely does not affect the decision of whether the variable ordered or categorical.
Note
If the dataset only contains input variables and no responses, use responseStartIdx = -2 and responseEndIdx = 0. The output variables vector will just contain zeros.

§ missingValue()

static float cv::ml::TrainData::missingValue ( )
inlinestatic

§ setTrainTestSplit()

virtual void cv::ml::TrainData::setTrainTestSplit ( int  count,
bool  shuffle = true 
)
pure virtual
Python:
None=cv.ml_TrainData.setTrainTestSplit(count[, shuffle])

Splits the training data into the training and test parts.

See also
TrainData::setTrainTestSplitRatio

§ setTrainTestSplitRatio()

virtual void cv::ml::TrainData::setTrainTestSplitRatio ( double  ratio,
bool  shuffle = true 
)
pure virtual
Python:
None=cv.ml_TrainData.setTrainTestSplitRatio(ratio[, shuffle])

Splits the training data into the training and test parts.

The function selects a subset of specified relative size and then returns it as the training set. If the function is not called, all the data is used for training. Please, note that for each of TrainData::getTrain* there is corresponding TrainData::getTest*, so that the test subset can be retrieved and processed as well.

See also
TrainData::setTrainTestSplit

§ shuffleTrainTest()

virtual void cv::ml::TrainData::shuffleTrainTest ( )
pure virtual
Python:
None=cv.ml_TrainData.shuffleTrainTest()

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