\section{All\+Categorical\+Split$<$ Fitness\+Function $>$ Class Template Reference} \label{classmlpack_1_1tree_1_1AllCategoricalSplit}\index{All\+Categorical\+Split$<$ Fitness\+Function $>$@{All\+Categorical\+Split$<$ Fitness\+Function $>$}} The \doxyref{All\+Categorical\+Split}{p.}{classmlpack_1_1tree_1_1AllCategoricalSplit} is a splitting function that will split categorical features into many children\+: one child for each category. \subsection*{Classes} \begin{DoxyCompactItemize} \item class \textbf{ Auxiliary\+Split\+Info} \end{DoxyCompactItemize} \subsection*{Static Public Member Functions} \begin{DoxyCompactItemize} \item {\footnotesize template$<$typename Elem\+Type $>$ }\\static size\+\_\+t \textbf{ Calculate\+Direction} (const Elem\+Type \&point, const arma\+::\+Col$<$ Elem\+Type $>$ \&class\+Probabilities, const \textbf{ Auxiliary\+Split\+Info}$<$ Elem\+Type $>$ \&) \begin{DoxyCompactList}\small\item\em Calculate the direction a point should percolate to. \end{DoxyCompactList}\item {\footnotesize template$<$typename Elem\+Type $>$ }\\static size\+\_\+t \textbf{ Num\+Children} (const arma\+::\+Col$<$ Elem\+Type $>$ \&class\+Probabilities, const \textbf{ Auxiliary\+Split\+Info}$<$ Elem\+Type $>$ \&) \begin{DoxyCompactList}\small\item\em Return the number of children in the split. \end{DoxyCompactList}\item {\footnotesize template$<$bool Use\+Weights, typename Vec\+Type , typename Weight\+Vec\+Type $>$ }\\static double \textbf{ Split\+If\+Better} (const double best\+Gain, const Vec\+Type \&data, const size\+\_\+t num\+Categories, const arma\+::\+Row$<$ size\+\_\+t $>$ \&labels, const size\+\_\+t num\+Classes, const Weight\+Vec\+Type \&weights, const size\+\_\+t minimum\+Leaf\+Size, const double minimum\+Gain\+Split, arma\+::\+Col$<$ typename Vec\+Type\+::elem\+\_\+type $>$ \&class\+Probabilities, \textbf{ Auxiliary\+Split\+Info}$<$ typename Vec\+Type\+::elem\+\_\+type $>$ \&aux) \begin{DoxyCompactList}\small\item\em Check if we can split a node. \end{DoxyCompactList}\end{DoxyCompactItemize} \subsection{Detailed Description} \subsubsection*{template$<$typename Fitness\+Function$>$\newline class mlpack\+::tree\+::\+All\+Categorical\+Split$<$ Fitness\+Function $>$} The \doxyref{All\+Categorical\+Split}{p.}{classmlpack_1_1tree_1_1AllCategoricalSplit} is a splitting function that will split categorical features into many children\+: one child for each category. \begin{DoxyTemplParams}{Template Parameters} {\em Fitness\+Function} & Fitness function to evaluate gain with. \\ \hline \end{DoxyTemplParams} Definition at line 28 of file all\+\_\+categorical\+\_\+split.\+hpp. \subsection{Member Function Documentation} \mbox{\label{classmlpack_1_1tree_1_1AllCategoricalSplit_adee8fec546042a57582d7aeab7425690}} \index{mlpack\+::tree\+::\+All\+Categorical\+Split@{mlpack\+::tree\+::\+All\+Categorical\+Split}!Calculate\+Direction@{Calculate\+Direction}} \index{Calculate\+Direction@{Calculate\+Direction}!mlpack\+::tree\+::\+All\+Categorical\+Split@{mlpack\+::tree\+::\+All\+Categorical\+Split}} \subsubsection{Calculate\+Direction()} {\footnotesize\ttfamily static size\+\_\+t Calculate\+Direction (\begin{DoxyParamCaption}\item[{const Elem\+Type \&}]{point, }\item[{const arma\+::\+Col$<$ Elem\+Type $>$ \&}]{class\+Probabilities, }\item[{const \textbf{ Auxiliary\+Split\+Info}$<$ Elem\+Type $>$ \&}]{ }\end{DoxyParamCaption})\hspace{0.3cm}{\ttfamily [static]}} Calculate the direction a point should percolate to. \begin{DoxyParams}{Parameters} {\em point} & the Point to use. \\ \hline {\em class\+Probabilities} & Column Vector of class probabilities. \\ \hline {\em $\ast$} & (aux) Auxiliary information for the split (Unused). \\ \hline \end{DoxyParams} \mbox{\label{classmlpack_1_1tree_1_1AllCategoricalSplit_adf571c6a85c385ffe99fe5a0b8322ef1}} \index{mlpack\+::tree\+::\+All\+Categorical\+Split@{mlpack\+::tree\+::\+All\+Categorical\+Split}!Num\+Children@{Num\+Children}} \index{Num\+Children@{Num\+Children}!mlpack\+::tree\+::\+All\+Categorical\+Split@{mlpack\+::tree\+::\+All\+Categorical\+Split}} \subsubsection{Num\+Children()} {\footnotesize\ttfamily static size\+\_\+t Num\+Children (\begin{DoxyParamCaption}\item[{const arma\+::\+Col$<$ Elem\+Type $>$ \&}]{class\+Probabilities, }\item[{const \textbf{ Auxiliary\+Split\+Info}$<$ Elem\+Type $>$ \&}]{ }\end{DoxyParamCaption})\hspace{0.3cm}{\ttfamily [static]}} Return the number of children in the split. \begin{DoxyParams}{Parameters} {\em class\+Probabilities} & Auxiliary information for the split. \\ \hline {\em $\ast$} & (aux) Auxiliary information for the split (Unused). \\ \hline \end{DoxyParams} \mbox{\label{classmlpack_1_1tree_1_1AllCategoricalSplit_ac03bd24ded8520b5e39a7023d3fb0ff0}} \index{mlpack\+::tree\+::\+All\+Categorical\+Split@{mlpack\+::tree\+::\+All\+Categorical\+Split}!Split\+If\+Better@{Split\+If\+Better}} \index{Split\+If\+Better@{Split\+If\+Better}!mlpack\+::tree\+::\+All\+Categorical\+Split@{mlpack\+::tree\+::\+All\+Categorical\+Split}} \subsubsection{Split\+If\+Better()} {\footnotesize\ttfamily static double Split\+If\+Better (\begin{DoxyParamCaption}\item[{const double}]{best\+Gain, }\item[{const Vec\+Type \&}]{data, }\item[{const size\+\_\+t}]{num\+Categories, }\item[{const arma\+::\+Row$<$ size\+\_\+t $>$ \&}]{labels, }\item[{const size\+\_\+t}]{num\+Classes, }\item[{const Weight\+Vec\+Type \&}]{weights, }\item[{const size\+\_\+t}]{minimum\+Leaf\+Size, }\item[{const double}]{minimum\+Gain\+Split, }\item[{arma\+::\+Col$<$ typename Vec\+Type\+::elem\+\_\+type $>$ \&}]{class\+Probabilities, }\item[{\textbf{ Auxiliary\+Split\+Info}$<$ typename Vec\+Type\+::elem\+\_\+type $>$ \&}]{aux }\end{DoxyParamCaption})\hspace{0.3cm}{\ttfamily [static]}} Check if we can split a node. If we can split a node in a way that improves on \textquotesingle{}best\+Gain\textquotesingle{}, then we return the improved gain. Otherwise we return the value \textquotesingle{}best\+Gain\textquotesingle{}. If a split is made, then class\+Probabilities and aux may be modified. For this particular split type, aux will be empty and class\+Probabilities will hold one element---the number of children. \begin{DoxyParams}{Parameters} {\em best\+Gain} & Best gain seen so far (we\textquotesingle{}ll only split if we find gain better than this). \\ \hline {\em data} & The dimension of data points to check for a split in. \\ \hline {\em num\+Categories} & Number of categories in the categorical data. \\ \hline {\em labels} & Labels for each point. \\ \hline {\em num\+Classes} & Number of classes in the dataset. \\ \hline {\em weights} & Weights associated with labels. \\ \hline {\em minimum\+Leaf\+Size} & Minimum number of points in a leaf node for splitting. \\ \hline {\em class\+Probabilities} & Class probabilities vector, which may be filled with split information a successful split. \\ \hline {\em minimum\+Gain\+Split} & Minimum gain split. \\ \hline {\em aux} & Auxiliary split information, which may be modified on a successful split. \\ \hline \end{DoxyParams} 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/decision\+\_\+tree/\textbf{ all\+\_\+categorical\+\_\+split.\+hpp}\end{DoxyCompactItemize}