GClasses
|
This uses graph-cut to divide the data into two clusters. It then trains a linear regression model for each cluster to map from inputs to change-in-state. It then aligns the smaller cluster with the larger one such that the linear models are in agreement (as much as possible). More...
#include <GManifold.h>
Public Member Functions | |
GDynamicSystemStateAligner (size_t neighbors, GMatrix &inputs, GRand &rand) | |
virtual | ~GDynamicSystemStateAligner () |
virtual GMatrix * | reduce (const GMatrix &in) |
Perform the transformation. More... | |
void | setSeeds (size_t a, size_t b) |
Specify the source and sink points for dividing the data into two clusters. More... | |
![]() | |
GTransform () | |
GTransform (GDomNode *pNode, GLearnerLoader &ll) | |
virtual | ~GTransform () |
Static Public Member Functions | |
static void | test () |
Protected Attributes | |
GMatrix & | m_inputs |
size_t | m_neighbors |
double * | m_pDistances |
size_t * | m_pNeighbors |
GRand & | m_rand |
size_t | m_seedA |
size_t | m_seedB |
Additional Inherited Members | |
![]() | |
virtual GDomNode * | baseDomNode (GDom *pDoc, const char *szClassName) const |
Child classes should use this in their implementation of serialize. More... | |
This uses graph-cut to divide the data into two clusters. It then trains a linear regression model for each cluster to map from inputs to change-in-state. It then aligns the smaller cluster with the larger one such that the linear models are in agreement (as much as possible).
GClasses::GDynamicSystemStateAligner::GDynamicSystemStateAligner | ( | size_t | neighbors, |
GMatrix & | inputs, | ||
GRand & | rand | ||
) |
|
virtual |
Perform the transformation.
Implements GClasses::GTransform.
void GClasses::GDynamicSystemStateAligner::setSeeds | ( | size_t | a, |
size_t | b | ||
) |
Specify the source and sink points for dividing the data into two clusters.
|
static |
|
protected |
|
protected |
|
protected |
|
protected |
|
protected |
|
protected |
|
protected |