Classifier Parameters. More...
#include <ClassifierEMStrategy.h>
Public Member Functions | |
| AbstractParameters * | clone () const |
| Create a clone copy of this instance. | |
| const Parameters & | operator= (const Parameters ¶ms) |
| Parameters () | |
| void | reset () |
| Clear all internal allocated resources and reset the parameters instance to its initial state. | |
| bool | serialize (AlgorithmParametersSerializer &serializer) const |
| Returns a parameter serialization object. | |
| ~Parameters () | |
Public Attributes | |
| std::vector< std::vector< double > > | m_clustersMeans |
| The previously estimated means of the clusters (optional). | |
| double | m_epsilon |
| The stop criteria. When the clusters change in a value smaller then epsilon, the convergence is achieved. | |
| unsigned int | m_maxInputPoints |
| The maximum number of points used to estimate the clusters (default = 1000). | |
| unsigned int | m_maxIterations |
| The maximum of iterations (E/M steps) to perform if convergence is not achieved. | |
| unsigned int | m_numberOfClusters |
| The number of clusters (classes) to estimate in the image. | |
| unsigned int | m_useRandomSamples |
| If true, random samples will be used instead of regular spaced samples. | |
Detailed Description
Definition at line 69 of file ClassifierEMStrategy.h.
Constructor & Destructor Documentation
◆ Parameters()
| te::rp::ClassifierEMStrategy::Parameters::Parameters | ( | ) |
◆ ~Parameters()
| te::rp::ClassifierEMStrategy::Parameters::~Parameters | ( | ) |
Member Function Documentation
◆ clone()
|
virtual |
Create a clone copy of this instance.
- Returns
- A clone copy of this instance.
- Note
- The caller will take the ownership of the returned pointer.
Implements te::common::AbstractParameters.
◆ operator=()
| const Parameters & te::rp::ClassifierEMStrategy::Parameters::operator= | ( | const Parameters & | params | ) |
◆ reset()
|
virtual |
Clear all internal allocated resources and reset the parameters instance to its initial state.
Implements te::common::AbstractParameters.
◆ serialize()
|
virtual |
Returns a parameter serialization object.
- Parameters
-
serializer The output serialization object.
- Returns
- Returns true if ok, false on erros.
Reimplemented from te::rp::ClassifierStrategyParameters.
Member Data Documentation
◆ m_clustersMeans
| std::vector<std::vector<double> > te::rp::ClassifierEMStrategy::Parameters::m_clustersMeans |
The previously estimated means of the clusters (optional).
Definition at line 78 of file ClassifierEMStrategy.h.
◆ m_epsilon
| double te::rp::ClassifierEMStrategy::Parameters::m_epsilon |
The stop criteria. When the clusters change in a value smaller then epsilon, the convergence is achieved.
Definition at line 77 of file ClassifierEMStrategy.h.
◆ m_maxInputPoints
| unsigned int te::rp::ClassifierEMStrategy::Parameters::m_maxInputPoints |
The maximum number of points used to estimate the clusters (default = 1000).
Definition at line 76 of file ClassifierEMStrategy.h.
◆ m_maxIterations
| unsigned int te::rp::ClassifierEMStrategy::Parameters::m_maxIterations |
The maximum of iterations (E/M steps) to perform if convergence is not achieved.
Definition at line 75 of file ClassifierEMStrategy.h.
◆ m_numberOfClusters
| unsigned int te::rp::ClassifierEMStrategy::Parameters::m_numberOfClusters |
The number of clusters (classes) to estimate in the image.
Definition at line 74 of file ClassifierEMStrategy.h.
◆ m_useRandomSamples
| unsigned int te::rp::ClassifierEMStrategy::Parameters::m_useRandomSamples |
If true, random samples will be used instead of regular spaced samples.
Definition at line 73 of file ClassifierEMStrategy.h.
The documentation for this class was generated from the following file:
- /home/castejon/castejon_files/develop/terralib5/git_release/src/terralib/rp/ClassifierEMStrategy.h
