An implementation of Neighborhood Components Analysis, both a linear dimensionality reduction technique and a distance learning technique. More...
Public Member Functions  
NCA (const arma::mat &dataset, const arma::Row< size_t > &labels, MetricType metric=MetricType())  
Construct the Neighborhood Components Analysis object. More...  
const arma::mat &  Dataset () const 
Get the dataset reference. More...  
const arma::Row< size_t > &  Labels () const 
Get the labels reference. More...  
void  LearnDistance (arma::mat &outputMatrix) 
Perform Neighborhood Components Analysis. More...  
const OptimizerType &  Optimizer () const 
Get the optimizer. More...  
OptimizerType &  Optimizer () 
An implementation of Neighborhood Components Analysis, both a linear dimensionality reduction technique and a distance learning technique.
The method seeks to improve knearestneighbor classification on a dataset by scaling the dimensions. The method is nonparametric, and does not require a value of k. It works by using stochastic ("soft") neighbor assignments and using optimization techniques over the gradient of the accuracy of the neighbor assignments.
For more details, see the following published paper:
NCA  (  const arma::mat &  dataset, 
const arma::Row< size_t > &  labels,  
MetricType  metric = MetricType() 

) 
Construct the Neighborhood Components Analysis object.
This simply stores the reference to the dataset and labels as well as the parameters for optimization before the actual optimization is performed.
dataset  Input dataset. 
labels  Input dataset labels. 
stepSize  Step size for stochastic gradient descent. 
maxIterations  Maximum iterations for stochastic gradient descent. 
tolerance  Tolerance for termination of stochastic gradient descent. 
shuffle  Whether or not to shuffle the dataset during SGD. 
metric  Instantiated metric to use. 

inline 

inline 
void LearnDistance  (  arma::mat &  outputMatrix  ) 
Perform Neighborhood Components Analysis.
The output distance learning matrix is written into the passed reference. If LearnDistance() is called with an outputMatrix which has the correct size (dataset.n_rows x dataset.n_rows), that matrix will be used as the starting point for optimization.
output_matrix  Covariance matrix of Mahalanobis distance. 

inline 