Machine Learning

Detailed Description

The Machine Learning Library (MLL) is a set of classes and functions for statistical classification, regression, and clustering of data.

Most of the classification and regression algorithms are implemented as C++ classes. As the algorithms have different sets of features (like an ability to handle missing measurements or categorical input variables), there is a little common ground between the classes. This common ground is defined by the class cv::ml::StatModel that all the other ML classes are derived from.

See detailed overview here: ml_intro.

Classes

Name	Description
class cv::ml::ANN_MLP	Artificial Neural Networks - Multi-Layer Perceptrons. View details
class cv::ml::Boost	Boosted tree classifier derived from DTrees. View details
class cv::ml::DTrees	The class represents a single decision tree or a collection of decision trees. View details
class cv::ml::EM	The class implements the Expectation Maximization algorithm. View details
class cv::ml::KNearest	The class implements K-Nearest Neighbors model. View details
class cv::ml::LogisticRegression	Implements Logistic Regression classifier. View details
class cv::ml::NormalBayesClassifier	Bayes classifier for normally distributed data. View details
class cv::ml::ParamGrid	The structure represents the logarithmic grid range of statmodel parameters. View details
class cv::ml::RTrees	The class implements the random forest predictor. View details
struct cv::ml::SimulatedAnnealingSolverSystem	This class declares example interface for system state used in simulated annealing optimization algorithm. View details
class cv::ml::StatModel	Base class for statistical models in OpenCV ML. View details
class cv::ml::SVM	Support Vector Machines. View details
class cv::ml::SVMSGD	Stochastic Gradient Descent SVM classifier. View details
class cv::ml::TrainData	Class encapsulating training data. View details

Enumerations

enum cv::ml::ErrorTypes {
    cv::ml::TEST_ERROR = 0,
    cv::ml::TRAIN_ERROR = 1
}

Error types View details

enum cv::ml::SampleTypes {
    cv::ml::ROW_SAMPLE = 0,
    cv::ml::COL_SAMPLE = 1
}

Sample types. View details

enum cv::ml::VariableTypes {
    cv::ml::VAR_NUMERICAL =0,
    cv::ml::VAR_ORDERED =0,
    cv::ml::VAR_CATEGORICAL =1
}

Variable types. View details

Typedef Documentation

ANN_MLP_ANNEAL

typedef ANN_MLP cv::ml::ANN_MLP_ANNEAL

#include <opencv2/ml.hpp>

Enumeration Type Documentation

ErrorTypes

enum cv::ml::ErrorTypes

#include <opencv2/ml.hpp>

Error types

Enumerator:

TEST_ERROR Python: cv.ml.TEST_ERROR

TRAIN_ERROR Python: cv.ml.TRAIN_ERROR

SampleTypes

enum cv::ml::SampleTypes

#include <opencv2/ml.hpp>

Sample types.

Enumerator:

ROW_SAMPLE Python: cv.ml.ROW_SAMPLE	each training sample is a row of samples
COL_SAMPLE Python: cv.ml.COL_SAMPLE	each training sample occupies a column of samples

VariableTypes

enum cv::ml::VariableTypes

#include <opencv2/ml.hpp>

Variable types.

Enumerator:

VAR_NUMERICAL Python: cv.ml.VAR_NUMERICAL	same as VAR_ORDERED
VAR_ORDERED Python: cv.ml.VAR_ORDERED	ordered variables
VAR_CATEGORICAL Python: cv.ml.VAR_CATEGORICAL	categorical variables

Function Documentation

createConcentricSpheresTestSet()

void cv::ml::createConcentricSpheresTestSet(
int nsamples,
int nfeatures,
int nclasses,
OutputArray samples,
OutputArray responses )

#include <opencv2/ml.hpp>

Creates test set.

Here is the call graph for this function:

randMVNormal()

void cv::ml::randMVNormal(
InputArray mean,
InputArray cov,
int nsamples,
OutputArray samples )

#include <opencv2/ml.hpp>

Generates sample from multivariate normal distribution.

Parameters

• mean — an average row vector

• cov — symmetric covariation matrix

• nsamples — returned samples count

• samples — returned samples array

Here is the call graph for this function:

simulatedAnnealingSolver()

template<class SimulatedAnnealingSolverSystem>
int cv::ml::simulatedAnnealingSolver(
SimulatedAnnealingSolverSystem & solverSystem,
double initialTemperature,
double finalTemperature,
double coolingRatio,
size_t iterationsPerStep,
double * lastTemperature = NULL,
cv::RNG & rngEnergy = cv::theRNG() )

#include <opencv2/ml.hpp>

The class implements simulated annealing for optimization.

[166] for details

Parameters

• solverSystem — optimization system (see SimulatedAnnealingSolverSystem)

• initialTemperature — initial temperature

• finalTemperature — final temperature

• coolingRatio — temperature step multiplies

• iterationsPerStep — number of iterations per temperature changing step

• lastTemperature — optional output for last used temperature

• rngEnergy — specify custom random numbers generator (cv::theRNG() by default)

Here is the call graph for this function: