\section{Sparse\+Coding Class Reference} \label{classmlpack_1_1sparse__coding_1_1SparseCoding}\index{Sparse\+Coding@{Sparse\+Coding}} An implementation of Sparse Coding with Dictionary Learning that achieves sparsity via an l1-\/norm regularizer on the codes (L\+A\+S\+SO) or an (l1+l2)-\/norm regularizer on the codes (the Elastic Net). \subsection*{Public Member Functions} \begin{DoxyCompactItemize} \item {\footnotesize template$<$typename Dictionary\+Initializer = Data\+Dependent\+Random\+Initializer$>$ }\\\textbf{ Sparse\+Coding} (const arma\+::mat \&data, const size\+\_\+t atoms, const double lambda1, const double lambda2=0, const size\+\_\+t max\+Iterations=0, const double obj\+Tolerance=0.\+01, const double newton\+Tolerance=1e-\/6, const Dictionary\+Initializer \&initializer=\+Dictionary\+Initializer()) \begin{DoxyCompactList}\small\item\em Set the parameters to \doxyref{Sparse\+Coding}{p.}{classmlpack_1_1sparse__coding_1_1SparseCoding}. \end{DoxyCompactList}\item \textbf{ Sparse\+Coding} (const size\+\_\+t atoms=0, const double lambda1=0, const double lambda2=0, const size\+\_\+t max\+Iterations=0, const double obj\+Tolerance=0.\+01, const double newton\+Tolerance=1e-\/6) \begin{DoxyCompactList}\small\item\em Set the parameters to \doxyref{Sparse\+Coding}{p.}{classmlpack_1_1sparse__coding_1_1SparseCoding}. \end{DoxyCompactList}\item size\+\_\+t \textbf{ Atoms} () const \begin{DoxyCompactList}\small\item\em Access the number of atoms. \end{DoxyCompactList}\item size\+\_\+t \& \textbf{ Atoms} () \begin{DoxyCompactList}\small\item\em Modify the number of atoms. \end{DoxyCompactList}\item const arma\+::mat \& \textbf{ Dictionary} () const \begin{DoxyCompactList}\small\item\em Access the dictionary. \end{DoxyCompactList}\item arma\+::mat \& \textbf{ Dictionary} () \begin{DoxyCompactList}\small\item\em Modify the dictionary. \end{DoxyCompactList}\item void \textbf{ Encode} (const arma\+::mat \&data, arma\+::mat \&codes) \begin{DoxyCompactList}\small\item\em Sparse code each point in the given dataset via L\+A\+RS, using the current dictionary and store the encoded data in the codes matrix. \end{DoxyCompactList}\item double \textbf{ Lambda1} () const \begin{DoxyCompactList}\small\item\em Access the L1 regularization term. \end{DoxyCompactList}\item double \& \textbf{ Lambda1} () \begin{DoxyCompactList}\small\item\em Modify the L1 regularization term. \end{DoxyCompactList}\item double \textbf{ Lambda2} () const \begin{DoxyCompactList}\small\item\em Access the L2 regularization term. \end{DoxyCompactList}\item double \& \textbf{ Lambda2} () \begin{DoxyCompactList}\small\item\em Modify the L2 regularization term. \end{DoxyCompactList}\item size\+\_\+t \textbf{ Max\+Iterations} () const \begin{DoxyCompactList}\small\item\em Get the maximum number of iterations. \end{DoxyCompactList}\item size\+\_\+t \& \textbf{ Max\+Iterations} () \begin{DoxyCompactList}\small\item\em Modify the maximum number of iterations. \end{DoxyCompactList}\item double \textbf{ Newton\+Tolerance} () const \begin{DoxyCompactList}\small\item\em Get the tolerance for Newton\textquotesingle{}s method (dictionary optimization step). \end{DoxyCompactList}\item double \& \textbf{ Newton\+Tolerance} () \begin{DoxyCompactList}\small\item\em Modify the tolerance for Newton\textquotesingle{}s method (dictionary optimization step). \end{DoxyCompactList}\item double \textbf{ Objective} (const arma\+::mat \&data, const arma\+::mat \&codes) const \begin{DoxyCompactList}\small\item\em Compute the objective function. \end{DoxyCompactList}\item double \textbf{ Obj\+Tolerance} () const \begin{DoxyCompactList}\small\item\em Get the objective tolerance. \end{DoxyCompactList}\item double \& \textbf{ Obj\+Tolerance} () \begin{DoxyCompactList}\small\item\em Modify the objective tolerance. \end{DoxyCompactList}\item double \textbf{ Optimize\+Dictionary} (const arma\+::mat \&data, const arma\+::mat \&codes, const arma\+::uvec \&adjacencies) \begin{DoxyCompactList}\small\item\em Learn dictionary via Newton method based on Lagrange dual. \end{DoxyCompactList}\item void \textbf{ Project\+Dictionary} () \begin{DoxyCompactList}\small\item\em Project each atom of the dictionary back onto the unit ball, if necessary. \end{DoxyCompactList}\item {\footnotesize template$<$typename Archive $>$ }\\void \textbf{ serialize} (Archive \&ar, const unsigned int) \begin{DoxyCompactList}\small\item\em Serialize the sparse coding model. \end{DoxyCompactList}\item {\footnotesize template$<$typename Dictionary\+Initializer = Data\+Dependent\+Random\+Initializer$>$ }\\double \textbf{ Train} (const arma\+::mat \&data, const Dictionary\+Initializer \&initializer=Dictionary\+Initializer()) \begin{DoxyCompactList}\small\item\em Train the sparse coding model on the given dataset. \end{DoxyCompactList}\end{DoxyCompactItemize} \subsection{Detailed Description} An implementation of Sparse Coding with Dictionary Learning that achieves sparsity via an l1-\/norm regularizer on the codes (L\+A\+S\+SO) or an (l1+l2)-\/norm regularizer on the codes (the Elastic Net). Let d be the number of dimensions in the original space, m the number of training points, and k the number of atoms in the dictionary (the dimension of the learned feature space). The training data X is a d-\/by-\/m matrix where each column is a point and each row is a dimension. The dictionary D is a d-\/by-\/k matrix, and the sparse codes matrix Z is a k-\/by-\/m matrix. This program seeks to minimize the objective\+: \[ \min_{D,Z} 0.5 ||X - D Z||_{F}^2\ + \lambda_1 \sum_{i=1}^m ||Z_i||_1 + 0.5 \lambda_2 \sum_{i=1}^m ||Z_i||_2^2 \] subject to $ ||D_j||_2 <= 1 $ for $ 1 <= j <= k $ where typically $ lambda_1 > 0 $ and $ lambda_2 = 0 $. This problem is solved by an algorithm that alternates between a dictionary learning step and a sparse coding step. The dictionary learning step updates the dictionary D using a Newton method based on the Lagrange dual (see the paper below for details). The sparse coding step involves solving a large number of sparse linear regression problems; this can be done efficiently using L\+A\+RS, an algorithm that can solve the L\+A\+S\+SO or the Elastic Net (papers below). Here are those papers\+: \begin{DoxyCode} @incollection\{lee2007efficient, title = \{Efficient sparse coding algorithms\}, author = \{Honglak Lee and Alexis Battle and Rajat Raina and Andrew Y. Ng\}, booktitle = \{Advances in Neural Information Processing Systems 19\}, editor = \{B. Sch\(\backslash\)\textcolor{stringliteral}{"\{o\}lkopf and J. Platt and T. Hoffman\},} \textcolor{stringliteral}{ publisher = \{MIT Press\},} \textcolor{stringliteral}{ address = \{Cambridge, MA\},} \textcolor{stringliteral}{ pages = \{801--808\},} \textcolor{stringliteral}{ year = \{2007\}} \textcolor{stringliteral}{\}} \end{DoxyCode} \begin{DoxyCode} @article\{efron2004least, title=\{Least angle regression\}, author=\{Efron, B. and Hastie, T. and Johnstone, I. and Tibshirani, R.\}, journal=\{The Annals of statistics\}, volume=\{32\}, number=\{2\}, pages=\{407--499\}, year=\{2004\}, publisher=\{Institute of Mathematical Statistics\} \} \end{DoxyCode} \begin{DoxyCode} @article\{zou2005regularization, title=\{Regularization and variable selection via the elastic net\}, author=\{Zou, H. and Hastie, T.\}, journal=\{Journal of the Royal Statistical Society Series B\}, volume=\{67\}, number=\{2\}, pages=\{301--320\}, year=\{2005\}, publisher=\{Royal Statistical Society\} \} \end{DoxyCode} Note that the implementation here does not use the feature-\/sign search algorithm from Honglak Lee\textquotesingle{}s paper, but instead the L\+A\+RS algorithm suggested in that paper. When \doxyref{Train()}{p.}{classmlpack_1_1sparse__coding_1_1SparseCoding_a69503a211de26a7b35e7e7867cf20246} is called, the dictionary is initialized using the Dictionary\+Initialization\+Policy class. Possible choices include the \doxyref{Random\+Initializer}{p.}{classmlpack_1_1sparse__coding_1_1RandomInitializer}, which provides an entirely random dictionary, the \doxyref{Data\+Dependent\+Random\+Initializer}{p.}{classmlpack_1_1sparse__coding_1_1DataDependentRandomInitializer}, which provides a random dictionary based loosely on characteristics of the dataset, and the \doxyref{Nothing\+Initializer}{p.}{classmlpack_1_1sparse__coding_1_1NothingInitializer}, which does not initialize the dictionary -- instead, the user should set the dictionary using the \doxyref{Dictionary()}{p.}{classmlpack_1_1sparse__coding_1_1SparseCoding_a3146526cfc85ff339121972d67c73f62} mutator method. Once a dictionary is trained with \doxyref{Train()}{p.}{classmlpack_1_1sparse__coding_1_1SparseCoding_a69503a211de26a7b35e7e7867cf20246}, another matrix may be encoded with the \doxyref{Encode()}{p.}{classmlpack_1_1sparse__coding_1_1SparseCoding_a4b2741b0e146d6eb6a09e30f228ea309} function. \begin{DoxyTemplParams}{Template Parameters} {\em Dictionary\+Initialization\+Policy} & The class to use to initialize the dictionary; must have \textquotesingle{}void Initialize(const arma\+::mat\& data, arma\+::mat\& dictionary)\textquotesingle{} function. \\ \hline \end{DoxyTemplParams} Definition at line 115 of file sparse\+\_\+coding.\+hpp. \subsection{Constructor \& Destructor Documentation} \mbox{\label{classmlpack_1_1sparse__coding_1_1SparseCoding_a603b4e6346773526111dd0e759ec1283}} \index{mlpack\+::sparse\+\_\+coding\+::\+Sparse\+Coding@{mlpack\+::sparse\+\_\+coding\+::\+Sparse\+Coding}!Sparse\+Coding@{Sparse\+Coding}} \index{Sparse\+Coding@{Sparse\+Coding}!mlpack\+::sparse\+\_\+coding\+::\+Sparse\+Coding@{mlpack\+::sparse\+\_\+coding\+::\+Sparse\+Coding}} \subsubsection{Sparse\+Coding()\hspace{0.1cm}{\footnotesize\ttfamily [1/2]}} {\footnotesize\ttfamily \textbf{ Sparse\+Coding} (\begin{DoxyParamCaption}\item[{const arma\+::mat \&}]{data, }\item[{const size\+\_\+t}]{atoms, }\item[{const double}]{lambda1, }\item[{const double}]{lambda2 = {\ttfamily 0}, }\item[{const size\+\_\+t}]{max\+Iterations = {\ttfamily 0}, }\item[{const double}]{obj\+Tolerance = {\ttfamily 0.01}, }\item[{const double}]{newton\+Tolerance = {\ttfamily 1e-\/6}, }\item[{const Dictionary\+Initializer \&}]{initializer = {\ttfamily DictionaryInitializer()} }\end{DoxyParamCaption})} Set the parameters to \doxyref{Sparse\+Coding}{p.}{classmlpack_1_1sparse__coding_1_1SparseCoding}. lambda2 defaults to 0. This constructor will train the model. If that is not desired, call the other constructor that does not take a data matrix. This constructor will also initialize the dictionary using the given Dictionary\+Initializer before training. If you want to initialize the dictionary to a custom matrix, consider either writing your own Dictionary\+Initializer class (with void Initialize(const arma\+::mat\& data, arma\+::mat\& dictionary) function), or call the constructor that does not take a data matrix, then call \doxyref{Dictionary()}{p.}{classmlpack_1_1sparse__coding_1_1SparseCoding_a3146526cfc85ff339121972d67c73f62} to set the dictionary matrix to a matrix of your choosing, and then call \doxyref{Train()}{p.}{classmlpack_1_1sparse__coding_1_1SparseCoding_a69503a211de26a7b35e7e7867cf20246} with \doxyref{Nothing\+Initializer}{p.}{classmlpack_1_1sparse__coding_1_1NothingInitializer} (i.\+e. Train$<$\+Nothing\+Initializer$>$(data)). \begin{DoxyParams}{Parameters} {\em data} & Data matrix. \\ \hline {\em atoms} & Number of atoms in dictionary. \\ \hline {\em lambda1} & Regularization parameter for l1-\/norm penalty. \\ \hline {\em lambda2} & Regularization parameter for l2-\/norm penalty. \\ \hline {\em max\+Iterations} & Maximum number of iterations to run algorithm. If 0, the algorithm will run until convergence (or forever). \\ \hline {\em obj\+Tolerance} & Tolerance for objective function. When an iteration of the algorithm produces an improvement smaller than this, the algorithm will terminate. \\ \hline {\em newton\+Tolerance} & Tolerance for the Newton\textquotesingle{}s method dictionary optimization step. \\ \hline {\em initializer} & The initializer to use. \\ \hline \end{DoxyParams} \mbox{\label{classmlpack_1_1sparse__coding_1_1SparseCoding_ad3476d9c5a489044fa1142e33a6052d7}} \index{mlpack\+::sparse\+\_\+coding\+::\+Sparse\+Coding@{mlpack\+::sparse\+\_\+coding\+::\+Sparse\+Coding}!Sparse\+Coding@{Sparse\+Coding}} \index{Sparse\+Coding@{Sparse\+Coding}!mlpack\+::sparse\+\_\+coding\+::\+Sparse\+Coding@{mlpack\+::sparse\+\_\+coding\+::\+Sparse\+Coding}} \subsubsection{Sparse\+Coding()\hspace{0.1cm}{\footnotesize\ttfamily [2/2]}} {\footnotesize\ttfamily \textbf{ Sparse\+Coding} (\begin{DoxyParamCaption}\item[{const size\+\_\+t}]{atoms = {\ttfamily 0}, }\item[{const double}]{lambda1 = {\ttfamily 0}, }\item[{const double}]{lambda2 = {\ttfamily 0}, }\item[{const size\+\_\+t}]{max\+Iterations = {\ttfamily 0}, }\item[{const double}]{obj\+Tolerance = {\ttfamily 0.01}, }\item[{const double}]{newton\+Tolerance = {\ttfamily 1e-\/6} }\end{DoxyParamCaption})} Set the parameters to \doxyref{Sparse\+Coding}{p.}{classmlpack_1_1sparse__coding_1_1SparseCoding}. lambda2 defaults to 0. This constructor will not train the model, and a subsequent call to \doxyref{Train()}{p.}{classmlpack_1_1sparse__coding_1_1SparseCoding_a69503a211de26a7b35e7e7867cf20246} will be required before the model can encode points with \doxyref{Encode()}{p.}{classmlpack_1_1sparse__coding_1_1SparseCoding_a4b2741b0e146d6eb6a09e30f228ea309}. \begin{DoxyParams}{Parameters} {\em atoms} & Number of atoms in dictionary. \\ \hline {\em lambda1} & Regularization parameter for l1-\/norm penalty. \\ \hline {\em lambda2} & Regularization parameter for l2-\/norm penalty. \\ \hline {\em max\+Iterations} & Maximum number of iterations to run algorithm. If 0, the algorithm will run until convergence (or forever). \\ \hline {\em obj\+Tolerance} & Tolerance for objective function. When an iteration of the algorithm produces an improvement smaller than this, the algorithm will terminate. \\ \hline {\em newton\+Tolerance} & Tolerance for the Newton\textquotesingle{}s method dictionary optimization step. \\ \hline \end{DoxyParams} \subsection{Member Function Documentation} \mbox{\label{classmlpack_1_1sparse__coding_1_1SparseCoding_ac1a924c2cc0b117283fc8aba76a92a07}} \index{mlpack\+::sparse\+\_\+coding\+::\+Sparse\+Coding@{mlpack\+::sparse\+\_\+coding\+::\+Sparse\+Coding}!Atoms@{Atoms}} \index{Atoms@{Atoms}!mlpack\+::sparse\+\_\+coding\+::\+Sparse\+Coding@{mlpack\+::sparse\+\_\+coding\+::\+Sparse\+Coding}} \subsubsection{Atoms()\hspace{0.1cm}{\footnotesize\ttfamily [1/2]}} {\footnotesize\ttfamily size\+\_\+t Atoms (\begin{DoxyParamCaption}{ }\end{DoxyParamCaption}) const\hspace{0.3cm}{\ttfamily [inline]}} Access the number of atoms. Definition at line 228 of file sparse\+\_\+coding.\+hpp. \mbox{\label{classmlpack_1_1sparse__coding_1_1SparseCoding_a30a05b58ad809476c03aba6df5970b52}} \index{mlpack\+::sparse\+\_\+coding\+::\+Sparse\+Coding@{mlpack\+::sparse\+\_\+coding\+::\+Sparse\+Coding}!Atoms@{Atoms}} \index{Atoms@{Atoms}!mlpack\+::sparse\+\_\+coding\+::\+Sparse\+Coding@{mlpack\+::sparse\+\_\+coding\+::\+Sparse\+Coding}} \subsubsection{Atoms()\hspace{0.1cm}{\footnotesize\ttfamily [2/2]}} {\footnotesize\ttfamily size\+\_\+t\& Atoms (\begin{DoxyParamCaption}{ }\end{DoxyParamCaption})\hspace{0.3cm}{\ttfamily [inline]}} Modify the number of atoms. Definition at line 230 of file sparse\+\_\+coding.\+hpp. \mbox{\label{classmlpack_1_1sparse__coding_1_1SparseCoding_a29bceb0b46c45a45594c9feb0e743171}} \index{mlpack\+::sparse\+\_\+coding\+::\+Sparse\+Coding@{mlpack\+::sparse\+\_\+coding\+::\+Sparse\+Coding}!Dictionary@{Dictionary}} \index{Dictionary@{Dictionary}!mlpack\+::sparse\+\_\+coding\+::\+Sparse\+Coding@{mlpack\+::sparse\+\_\+coding\+::\+Sparse\+Coding}} \subsubsection{Dictionary()\hspace{0.1cm}{\footnotesize\ttfamily [1/2]}} {\footnotesize\ttfamily const arma\+::mat\& Dictionary (\begin{DoxyParamCaption}{ }\end{DoxyParamCaption}) const\hspace{0.3cm}{\ttfamily [inline]}} Access the dictionary. Definition at line 223 of file sparse\+\_\+coding.\+hpp. \mbox{\label{classmlpack_1_1sparse__coding_1_1SparseCoding_a3146526cfc85ff339121972d67c73f62}} \index{mlpack\+::sparse\+\_\+coding\+::\+Sparse\+Coding@{mlpack\+::sparse\+\_\+coding\+::\+Sparse\+Coding}!Dictionary@{Dictionary}} \index{Dictionary@{Dictionary}!mlpack\+::sparse\+\_\+coding\+::\+Sparse\+Coding@{mlpack\+::sparse\+\_\+coding\+::\+Sparse\+Coding}} \subsubsection{Dictionary()\hspace{0.1cm}{\footnotesize\ttfamily [2/2]}} {\footnotesize\ttfamily arma\+::mat\& Dictionary (\begin{DoxyParamCaption}{ }\end{DoxyParamCaption})\hspace{0.3cm}{\ttfamily [inline]}} Modify the dictionary. Definition at line 225 of file sparse\+\_\+coding.\+hpp. \mbox{\label{classmlpack_1_1sparse__coding_1_1SparseCoding_a4b2741b0e146d6eb6a09e30f228ea309}} \index{mlpack\+::sparse\+\_\+coding\+::\+Sparse\+Coding@{mlpack\+::sparse\+\_\+coding\+::\+Sparse\+Coding}!Encode@{Encode}} \index{Encode@{Encode}!mlpack\+::sparse\+\_\+coding\+::\+Sparse\+Coding@{mlpack\+::sparse\+\_\+coding\+::\+Sparse\+Coding}} \subsubsection{Encode()} {\footnotesize\ttfamily void Encode (\begin{DoxyParamCaption}\item[{const arma\+::mat \&}]{data, }\item[{arma\+::mat \&}]{codes }\end{DoxyParamCaption})} Sparse code each point in the given dataset via L\+A\+RS, using the current dictionary and store the encoded data in the codes matrix. \begin{DoxyParams}{Parameters} {\em data} & Input data matrix to be encoded. \\ \hline {\em codes} & Output codes matrix. \\ \hline \end{DoxyParams} \mbox{\label{classmlpack_1_1sparse__coding_1_1SparseCoding_a264615b0081adef98427420db074cd44}} \index{mlpack\+::sparse\+\_\+coding\+::\+Sparse\+Coding@{mlpack\+::sparse\+\_\+coding\+::\+Sparse\+Coding}!Lambda1@{Lambda1}} \index{Lambda1@{Lambda1}!mlpack\+::sparse\+\_\+coding\+::\+Sparse\+Coding@{mlpack\+::sparse\+\_\+coding\+::\+Sparse\+Coding}} \subsubsection{Lambda1()\hspace{0.1cm}{\footnotesize\ttfamily [1/2]}} {\footnotesize\ttfamily double Lambda1 (\begin{DoxyParamCaption}{ }\end{DoxyParamCaption}) const\hspace{0.3cm}{\ttfamily [inline]}} Access the L1 regularization term. Definition at line 233 of file sparse\+\_\+coding.\+hpp. \mbox{\label{classmlpack_1_1sparse__coding_1_1SparseCoding_a743296069b1e03a1bb2e7a466eb68022}} \index{mlpack\+::sparse\+\_\+coding\+::\+Sparse\+Coding@{mlpack\+::sparse\+\_\+coding\+::\+Sparse\+Coding}!Lambda1@{Lambda1}} \index{Lambda1@{Lambda1}!mlpack\+::sparse\+\_\+coding\+::\+Sparse\+Coding@{mlpack\+::sparse\+\_\+coding\+::\+Sparse\+Coding}} \subsubsection{Lambda1()\hspace{0.1cm}{\footnotesize\ttfamily [2/2]}} {\footnotesize\ttfamily double\& Lambda1 (\begin{DoxyParamCaption}{ }\end{DoxyParamCaption})\hspace{0.3cm}{\ttfamily [inline]}} Modify the L1 regularization term. Definition at line 235 of file sparse\+\_\+coding.\+hpp. \mbox{\label{classmlpack_1_1sparse__coding_1_1SparseCoding_aae12d0e36b1a7fa52a155f9d3f7a50f1}} \index{mlpack\+::sparse\+\_\+coding\+::\+Sparse\+Coding@{mlpack\+::sparse\+\_\+coding\+::\+Sparse\+Coding}!Lambda2@{Lambda2}} \index{Lambda2@{Lambda2}!mlpack\+::sparse\+\_\+coding\+::\+Sparse\+Coding@{mlpack\+::sparse\+\_\+coding\+::\+Sparse\+Coding}} \subsubsection{Lambda2()\hspace{0.1cm}{\footnotesize\ttfamily [1/2]}} {\footnotesize\ttfamily double Lambda2 (\begin{DoxyParamCaption}{ }\end{DoxyParamCaption}) const\hspace{0.3cm}{\ttfamily [inline]}} Access the L2 regularization term. Definition at line 238 of file sparse\+\_\+coding.\+hpp. \mbox{\label{classmlpack_1_1sparse__coding_1_1SparseCoding_adb32f19e01e1576af72865eb87978008}} \index{mlpack\+::sparse\+\_\+coding\+::\+Sparse\+Coding@{mlpack\+::sparse\+\_\+coding\+::\+Sparse\+Coding}!Lambda2@{Lambda2}} \index{Lambda2@{Lambda2}!mlpack\+::sparse\+\_\+coding\+::\+Sparse\+Coding@{mlpack\+::sparse\+\_\+coding\+::\+Sparse\+Coding}} \subsubsection{Lambda2()\hspace{0.1cm}{\footnotesize\ttfamily [2/2]}} {\footnotesize\ttfamily double\& Lambda2 (\begin{DoxyParamCaption}{ }\end{DoxyParamCaption})\hspace{0.3cm}{\ttfamily [inline]}} Modify the L2 regularization term. Definition at line 240 of file sparse\+\_\+coding.\+hpp. \mbox{\label{classmlpack_1_1sparse__coding_1_1SparseCoding_a420770944a5b0c7a852c4ec372c4a2d1}} \index{mlpack\+::sparse\+\_\+coding\+::\+Sparse\+Coding@{mlpack\+::sparse\+\_\+coding\+::\+Sparse\+Coding}!Max\+Iterations@{Max\+Iterations}} \index{Max\+Iterations@{Max\+Iterations}!mlpack\+::sparse\+\_\+coding\+::\+Sparse\+Coding@{mlpack\+::sparse\+\_\+coding\+::\+Sparse\+Coding}} \subsubsection{Max\+Iterations()\hspace{0.1cm}{\footnotesize\ttfamily [1/2]}} {\footnotesize\ttfamily size\+\_\+t Max\+Iterations (\begin{DoxyParamCaption}{ }\end{DoxyParamCaption}) const\hspace{0.3cm}{\ttfamily [inline]}} Get the maximum number of iterations. Definition at line 243 of file sparse\+\_\+coding.\+hpp. \mbox{\label{classmlpack_1_1sparse__coding_1_1SparseCoding_acda675ab4ab86b95c92bc33bc391a61b}} \index{mlpack\+::sparse\+\_\+coding\+::\+Sparse\+Coding@{mlpack\+::sparse\+\_\+coding\+::\+Sparse\+Coding}!Max\+Iterations@{Max\+Iterations}} \index{Max\+Iterations@{Max\+Iterations}!mlpack\+::sparse\+\_\+coding\+::\+Sparse\+Coding@{mlpack\+::sparse\+\_\+coding\+::\+Sparse\+Coding}} \subsubsection{Max\+Iterations()\hspace{0.1cm}{\footnotesize\ttfamily [2/2]}} {\footnotesize\ttfamily size\+\_\+t\& Max\+Iterations (\begin{DoxyParamCaption}{ }\end{DoxyParamCaption})\hspace{0.3cm}{\ttfamily [inline]}} Modify the maximum number of iterations. Definition at line 245 of file sparse\+\_\+coding.\+hpp. \mbox{\label{classmlpack_1_1sparse__coding_1_1SparseCoding_a6fcf2b9f21e7933d9292d11be1535475}} \index{mlpack\+::sparse\+\_\+coding\+::\+Sparse\+Coding@{mlpack\+::sparse\+\_\+coding\+::\+Sparse\+Coding}!Newton\+Tolerance@{Newton\+Tolerance}} \index{Newton\+Tolerance@{Newton\+Tolerance}!mlpack\+::sparse\+\_\+coding\+::\+Sparse\+Coding@{mlpack\+::sparse\+\_\+coding\+::\+Sparse\+Coding}} \subsubsection{Newton\+Tolerance()\hspace{0.1cm}{\footnotesize\ttfamily [1/2]}} {\footnotesize\ttfamily double Newton\+Tolerance (\begin{DoxyParamCaption}{ }\end{DoxyParamCaption}) const\hspace{0.3cm}{\ttfamily [inline]}} Get the tolerance for Newton\textquotesingle{}s method (dictionary optimization step). Definition at line 253 of file sparse\+\_\+coding.\+hpp. \mbox{\label{classmlpack_1_1sparse__coding_1_1SparseCoding_a32f233437346e49c0b7dad746ac62b6b}} \index{mlpack\+::sparse\+\_\+coding\+::\+Sparse\+Coding@{mlpack\+::sparse\+\_\+coding\+::\+Sparse\+Coding}!Newton\+Tolerance@{Newton\+Tolerance}} \index{Newton\+Tolerance@{Newton\+Tolerance}!mlpack\+::sparse\+\_\+coding\+::\+Sparse\+Coding@{mlpack\+::sparse\+\_\+coding\+::\+Sparse\+Coding}} \subsubsection{Newton\+Tolerance()\hspace{0.1cm}{\footnotesize\ttfamily [2/2]}} {\footnotesize\ttfamily double\& Newton\+Tolerance (\begin{DoxyParamCaption}{ }\end{DoxyParamCaption})\hspace{0.3cm}{\ttfamily [inline]}} Modify the tolerance for Newton\textquotesingle{}s method (dictionary optimization step). Definition at line 255 of file sparse\+\_\+coding.\+hpp. References Sparse\+Coding\+::serialize(). \mbox{\label{classmlpack_1_1sparse__coding_1_1SparseCoding_a8a45b3d947cc22079a5941b9d46acfe3}} \index{mlpack\+::sparse\+\_\+coding\+::\+Sparse\+Coding@{mlpack\+::sparse\+\_\+coding\+::\+Sparse\+Coding}!Objective@{Objective}} \index{Objective@{Objective}!mlpack\+::sparse\+\_\+coding\+::\+Sparse\+Coding@{mlpack\+::sparse\+\_\+coding\+::\+Sparse\+Coding}} \subsubsection{Objective()} {\footnotesize\ttfamily double Objective (\begin{DoxyParamCaption}\item[{const arma\+::mat \&}]{data, }\item[{const arma\+::mat \&}]{codes }\end{DoxyParamCaption}) const} Compute the objective function. \mbox{\label{classmlpack_1_1sparse__coding_1_1SparseCoding_a8c2c8dec8e93ca1ddc14ac238266ef20}} \index{mlpack\+::sparse\+\_\+coding\+::\+Sparse\+Coding@{mlpack\+::sparse\+\_\+coding\+::\+Sparse\+Coding}!Obj\+Tolerance@{Obj\+Tolerance}} \index{Obj\+Tolerance@{Obj\+Tolerance}!mlpack\+::sparse\+\_\+coding\+::\+Sparse\+Coding@{mlpack\+::sparse\+\_\+coding\+::\+Sparse\+Coding}} \subsubsection{Obj\+Tolerance()\hspace{0.1cm}{\footnotesize\ttfamily [1/2]}} {\footnotesize\ttfamily double Obj\+Tolerance (\begin{DoxyParamCaption}{ }\end{DoxyParamCaption}) const\hspace{0.3cm}{\ttfamily [inline]}} Get the objective tolerance. Definition at line 248 of file sparse\+\_\+coding.\+hpp. \mbox{\label{classmlpack_1_1sparse__coding_1_1SparseCoding_afe546afaa975282a5549cc416c1d64f5}} \index{mlpack\+::sparse\+\_\+coding\+::\+Sparse\+Coding@{mlpack\+::sparse\+\_\+coding\+::\+Sparse\+Coding}!Obj\+Tolerance@{Obj\+Tolerance}} \index{Obj\+Tolerance@{Obj\+Tolerance}!mlpack\+::sparse\+\_\+coding\+::\+Sparse\+Coding@{mlpack\+::sparse\+\_\+coding\+::\+Sparse\+Coding}} \subsubsection{Obj\+Tolerance()\hspace{0.1cm}{\footnotesize\ttfamily [2/2]}} {\footnotesize\ttfamily double\& Obj\+Tolerance (\begin{DoxyParamCaption}{ }\end{DoxyParamCaption})\hspace{0.3cm}{\ttfamily [inline]}} Modify the objective tolerance. Definition at line 250 of file sparse\+\_\+coding.\+hpp. \mbox{\label{classmlpack_1_1sparse__coding_1_1SparseCoding_a2d164dea45075388335d75f8c27fd659}} \index{mlpack\+::sparse\+\_\+coding\+::\+Sparse\+Coding@{mlpack\+::sparse\+\_\+coding\+::\+Sparse\+Coding}!Optimize\+Dictionary@{Optimize\+Dictionary}} \index{Optimize\+Dictionary@{Optimize\+Dictionary}!mlpack\+::sparse\+\_\+coding\+::\+Sparse\+Coding@{mlpack\+::sparse\+\_\+coding\+::\+Sparse\+Coding}} \subsubsection{Optimize\+Dictionary()} {\footnotesize\ttfamily double Optimize\+Dictionary (\begin{DoxyParamCaption}\item[{const arma\+::mat \&}]{data, }\item[{const arma\+::mat \&}]{codes, }\item[{const arma\+::uvec \&}]{adjacencies }\end{DoxyParamCaption})} Learn dictionary via Newton method based on Lagrange dual. \begin{DoxyParams}{Parameters} {\em data} & Data matrix. \\ \hline {\em codes} & Matrix of codes. \\ \hline {\em adjacencies} & Indices of entries (unrolled column by column) of the coding matrix Z that are non-\/zero (the adjacency matrix for the bipartite graph of points and atoms). \\ \hline \end{DoxyParams} \begin{DoxyReturn}{Returns} the norm of the gradient of the Lagrange dual with respect to the dual variables \end{DoxyReturn} \mbox{\label{classmlpack_1_1sparse__coding_1_1SparseCoding_ab8a90e33cb1a5db9cda2ef2f0ba26e5f}} \index{mlpack\+::sparse\+\_\+coding\+::\+Sparse\+Coding@{mlpack\+::sparse\+\_\+coding\+::\+Sparse\+Coding}!Project\+Dictionary@{Project\+Dictionary}} \index{Project\+Dictionary@{Project\+Dictionary}!mlpack\+::sparse\+\_\+coding\+::\+Sparse\+Coding@{mlpack\+::sparse\+\_\+coding\+::\+Sparse\+Coding}} \subsubsection{Project\+Dictionary()} {\footnotesize\ttfamily void Project\+Dictionary (\begin{DoxyParamCaption}{ }\end{DoxyParamCaption})} Project each atom of the dictionary back onto the unit ball, if necessary. \mbox{\label{classmlpack_1_1sparse__coding_1_1SparseCoding_af0dd9205158ccf7bcfcd8ff81f79c927}} \index{mlpack\+::sparse\+\_\+coding\+::\+Sparse\+Coding@{mlpack\+::sparse\+\_\+coding\+::\+Sparse\+Coding}!serialize@{serialize}} \index{serialize@{serialize}!mlpack\+::sparse\+\_\+coding\+::\+Sparse\+Coding@{mlpack\+::sparse\+\_\+coding\+::\+Sparse\+Coding}} \subsubsection{serialize()} {\footnotesize\ttfamily void serialize (\begin{DoxyParamCaption}\item[{Archive \&}]{ar, }\item[{const unsigned}]{int }\end{DoxyParamCaption})} Serialize the sparse coding model. Referenced by Sparse\+Coding\+::\+Newton\+Tolerance(). \mbox{\label{classmlpack_1_1sparse__coding_1_1SparseCoding_a69503a211de26a7b35e7e7867cf20246}} \index{mlpack\+::sparse\+\_\+coding\+::\+Sparse\+Coding@{mlpack\+::sparse\+\_\+coding\+::\+Sparse\+Coding}!Train@{Train}} \index{Train@{Train}!mlpack\+::sparse\+\_\+coding\+::\+Sparse\+Coding@{mlpack\+::sparse\+\_\+coding\+::\+Sparse\+Coding}} \subsubsection{Train()} {\footnotesize\ttfamily double Train (\begin{DoxyParamCaption}\item[{const arma\+::mat \&}]{data, }\item[{const Dictionary\+Initializer \&}]{initializer = {\ttfamily DictionaryInitializer()} }\end{DoxyParamCaption})} Train the sparse coding model on the given dataset. \begin{DoxyReturn}{Returns} The final objective value. \end{DoxyReturn} The documentation for this class was generated from the following file\+:\begin{DoxyCompactItemize} \item /var/www/mlpack.\+ratml.\+org/mlpack.\+org/\+\_\+src/mlpack-\/3.\+3.\+2/src/mlpack/methods/sparse\+\_\+coding/\textbf{ sparse\+\_\+coding.\+hpp}\end{DoxyCompactItemize}