#include <cluster.h>
Public Methods | |
| ClusterGauss (Vector< float > ¢er, float alpha=1.0) | |
| ClusterGauss (Vector< float > ¢er, Matrix< float > &covariance, float alpha=1.0) | |
| ClusterGauss (std::istream &stream) | |
| Construct from stream. | |
| ~ClusterGauss () | |
| void | prepareInverse () |
| When covariance is changed, update cached information necessary to compute Mahalanobis distance. | |
| float | probability (const Vector< float > &point, float *scale=NULL, float *minScale=NULL) |
| The probability of being in the cluster, which is simply the Gaussian of the distance from the center. Result is multiplied by exp (scale) if minScale == NULL; otherwise scale and minScale are updated, and result is unscaled. | |
| void | read (std::istream &stream) |
| void | write (std::ostream &stream) |
Public Attributes | |
| float | alpha |
| Vector< float > | center |
| Matrix< float > | covariance |
| Matrix< float > | eigenvectors |
| Vector< float > | eigenvalues |
| Matrix< float > | eigenverse |
| float | det |
| preprocessed multiplier that goes in front of probability expression. Includes determinant of the covariance matrix. | |
|
||||||||||||
|
|
|
||||||||||||||||
|
|
|
|
Construct from stream.
|
|
|
|
|
|
When covariance is changed, update cached information necessary to compute Mahalanobis distance.
|
|
||||||||||||||||
|
The probability of being in the cluster, which is simply the Gaussian of the distance from the center. Result is multiplied by exp (scale) if minScale == NULL; otherwise scale and minScale are updated, and result is unscaled.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
preprocessed multiplier that goes in front of probability expression. Includes determinant of the covariance matrix.
|
|
|
|
|
|
|
|
|
|
1.2.18