\section{N\+CA$<$ Metric\+Type, Optimizer\+Type $>$ Class Template Reference}
\label{classmlpack_1_1nca_1_1NCA}\index{N\+C\+A$<$ Metric\+Type, Optimizer\+Type $>$@{N\+C\+A$<$ Metric\+Type, Optimizer\+Type $>$}}


An implementation of Neighborhood Components Analysis, both a linear dimensionality reduction technique and a distance learning technique.  


\subsection*{Public Member Functions}
\begin{DoxyCompactItemize}
\item 
\textbf{ N\+CA} (const arma\+::mat \&dataset, const arma\+::\+Row$<$ size\+\_\+t $>$ \&labels, Metric\+Type metric=Metric\+Type())
\begin{DoxyCompactList}\small\item\em Construct the Neighborhood Components Analysis object. \end{DoxyCompactList}\item 
const arma\+::mat \& \textbf{ Dataset} () const
\begin{DoxyCompactList}\small\item\em Get the dataset reference. \end{DoxyCompactList}\item 
const arma\+::\+Row$<$ size\+\_\+t $>$ \& \textbf{ Labels} () const
\begin{DoxyCompactList}\small\item\em Get the labels reference. \end{DoxyCompactList}\item 
{\footnotesize template$<$typename... Callback\+Types$>$ }\\void \textbf{ Learn\+Distance} (arma\+::mat \&output\+Matrix, Callback\+Types \&\&... callbacks)
\begin{DoxyCompactList}\small\item\em Perform Neighborhood Components Analysis. \end{DoxyCompactList}\item 
const Optimizer\+Type \& \textbf{ Optimizer} () const
\begin{DoxyCompactList}\small\item\em Get the optimizer. \end{DoxyCompactList}\item 
Optimizer\+Type \& \textbf{ Optimizer} ()
\end{DoxyCompactItemize}


\subsection{Detailed Description}
\subsubsection*{template$<$typename Metric\+Type = metric\+::\+Squared\+Euclidean\+Distance, typename Optimizer\+Type = ens\+::\+Standard\+S\+GD$>$\newline
class mlpack\+::nca\+::\+N\+C\+A$<$ Metric\+Type, Optimizer\+Type $>$}

An implementation of Neighborhood Components Analysis, both a linear dimensionality reduction technique and a distance learning technique. 

The method seeks to improve k-\/nearest-\/neighbor 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 (\char`\"{}soft\char`\"{}) neighbor assignments and using optimization techniques over the gradient of the accuracy of the neighbor assignments.

For more details, see the following published paper\+:


\begin{DoxyCode}
@inproceedings\{Goldberger2004,
  author = \{Goldberger, Jacob and Roweis, Sam and Hinton, Geoff and
      Salakhutdinov, Ruslan\},
  booktitle = \{Advances in Neural Information Processing Systems 17\},
  pages = \{513--520\},
  publisher = \{MIT Press\},
  title = \{\{Neighbourhood Components Analysis\}\},
  year = \{2004\}
\}
\end{DoxyCode}
 

Definition at line 49 of file nca.\+hpp.



\subsection{Constructor \& Destructor Documentation}
\mbox{\label{classmlpack_1_1nca_1_1NCA_a8a7a5b0fbf5a82acaf0e316944998f63}} 
\index{mlpack\+::nca\+::\+N\+CA@{mlpack\+::nca\+::\+N\+CA}!N\+CA@{N\+CA}}
\index{N\+CA@{N\+CA}!mlpack\+::nca\+::\+N\+CA@{mlpack\+::nca\+::\+N\+CA}}
\subsubsection{N\+C\+A()}
{\footnotesize\ttfamily \textbf{ N\+CA} (\begin{DoxyParamCaption}\item[{const arma\+::mat \&}]{dataset,  }\item[{const arma\+::\+Row$<$ size\+\_\+t $>$ \&}]{labels,  }\item[{Metric\+Type}]{metric = {\ttfamily MetricType()} }\end{DoxyParamCaption})}



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.


\begin{DoxyParams}{Parameters}
{\em dataset} & Input dataset. \\
\hline
{\em labels} & Input dataset labels. \\
\hline
{\em metric} & Instantiated metric to use. \\
\hline
\end{DoxyParams}


\subsection{Member Function Documentation}
\mbox{\label{classmlpack_1_1nca_1_1NCA_aff320b9a86b77a150e630c01d5888273}} 
\index{mlpack\+::nca\+::\+N\+CA@{mlpack\+::nca\+::\+N\+CA}!Dataset@{Dataset}}
\index{Dataset@{Dataset}!mlpack\+::nca\+::\+N\+CA@{mlpack\+::nca\+::\+N\+CA}}
\subsubsection{Dataset()}
{\footnotesize\ttfamily const arma\+::mat\& Dataset (\begin{DoxyParamCaption}{ }\end{DoxyParamCaption}) const\hspace{0.3cm}{\ttfamily [inline]}}



Get the dataset reference. 



Definition at line 81 of file nca.\+hpp.

\mbox{\label{classmlpack_1_1nca_1_1NCA_ae68718a757e9635f87377fd3935ca20d}} 
\index{mlpack\+::nca\+::\+N\+CA@{mlpack\+::nca\+::\+N\+CA}!Labels@{Labels}}
\index{Labels@{Labels}!mlpack\+::nca\+::\+N\+CA@{mlpack\+::nca\+::\+N\+CA}}
\subsubsection{Labels()}
{\footnotesize\ttfamily const arma\+::\+Row$<$size\+\_\+t$>$\& Labels (\begin{DoxyParamCaption}{ }\end{DoxyParamCaption}) const\hspace{0.3cm}{\ttfamily [inline]}}



Get the labels reference. 



Definition at line 83 of file nca.\+hpp.

\mbox{\label{classmlpack_1_1nca_1_1NCA_a55126379a6e9673a59546f4e446e9ada}} 
\index{mlpack\+::nca\+::\+N\+CA@{mlpack\+::nca\+::\+N\+CA}!Learn\+Distance@{Learn\+Distance}}
\index{Learn\+Distance@{Learn\+Distance}!mlpack\+::nca\+::\+N\+CA@{mlpack\+::nca\+::\+N\+CA}}
\subsubsection{Learn\+Distance()}
{\footnotesize\ttfamily void Learn\+Distance (\begin{DoxyParamCaption}\item[{arma\+::mat \&}]{output\+Matrix,  }\item[{Callback\+Types \&\&...}]{callbacks }\end{DoxyParamCaption})}



Perform Neighborhood Components Analysis. 

The output distance learning matrix is written into the passed reference. If \doxyref{Learn\+Distance()}{p.}{classmlpack_1_1nca_1_1NCA_a55126379a6e9673a59546f4e446e9ada} is called with an output\+Matrix which has the correct size (dataset.\+n\+\_\+rows x dataset.\+n\+\_\+rows), that matrix will be used as the starting point for optimization.


\begin{DoxyTemplParams}{Template Parameters}
{\em Callback\+Types} & Types of Callback functions. \\
\hline
\end{DoxyTemplParams}

\begin{DoxyParams}{Parameters}
{\em output\+Matrix} & Covariance matrix of Mahalanobis distance. \\
\hline
{\em callbacks} & Callback function for ensmallen optimizer {\ttfamily Optimizer\+Type}. See {\tt https\+://www.\+ensmallen.\+org/docs.\+html\#callback-\/documentation}. \\
\hline
\end{DoxyParams}
\mbox{\label{classmlpack_1_1nca_1_1NCA_a3bdf808beacdb2f464caa78076a23a0a}} 
\index{mlpack\+::nca\+::\+N\+CA@{mlpack\+::nca\+::\+N\+CA}!Optimizer@{Optimizer}}
\index{Optimizer@{Optimizer}!mlpack\+::nca\+::\+N\+CA@{mlpack\+::nca\+::\+N\+CA}}
\subsubsection{Optimizer()\hspace{0.1cm}{\footnotesize\ttfamily [1/2]}}
{\footnotesize\ttfamily const Optimizer\+Type\& Optimizer (\begin{DoxyParamCaption}{ }\end{DoxyParamCaption}) const\hspace{0.3cm}{\ttfamily [inline]}}



Get the optimizer. 



Definition at line 86 of file nca.\+hpp.

\mbox{\label{classmlpack_1_1nca_1_1NCA_a213f61b8936eadf8a162e45b4c6188d0}} 
\index{mlpack\+::nca\+::\+N\+CA@{mlpack\+::nca\+::\+N\+CA}!Optimizer@{Optimizer}}
\index{Optimizer@{Optimizer}!mlpack\+::nca\+::\+N\+CA@{mlpack\+::nca\+::\+N\+CA}}
\subsubsection{Optimizer()\hspace{0.1cm}{\footnotesize\ttfamily [2/2]}}
{\footnotesize\ttfamily Optimizer\+Type\& Optimizer (\begin{DoxyParamCaption}{ }\end{DoxyParamCaption})\hspace{0.3cm}{\ttfamily [inline]}}



Definition at line 87 of file nca.\+hpp.



The documentation for this class was generated from the following file\+:\begin{DoxyCompactItemize}
\item 
/var/www/mlpack.\+ratml.\+org/mlpack.\+org/\+\_\+src/mlpack-\/git/src/mlpack/methods/nca/\textbf{ nca.\+hpp}\end{DoxyCompactItemize}