Description of sladjmat

Home > sltoolbox > graph > sladjmat.m

sladjmat

PURPOSE

SLADJMAT Constructs the adjacency matrix representation of a graph

SYNOPSIS

function A = sladjmat(G, varargin)

DESCRIPTION

SLADJMAT Constructs the adjacency matrix representation of a graph

 $ Syntax $
   - A = sladjmat(G, ...)

 $ Arguments $
   - G:        The input graph
   - A:        The adjacency matrix representation of the graph

 $ Description $
   - A = sladjmat(G, ...) constructs the adjacency matrix representation
     of a graph. You can specify the following properties to control
     the construction.
       - 'valtype':        the value type of the target matrix
                           - 'auto': if has value, make numeric matrix
                                     if no value, make logical matrix
                           - 'logical': make logical matrix always
                           - 'numeric': make numeric matrix always
                           (default = 'auto')
       - 'sparse':         whether to create a sparse matrix 
                           (default = true)
       - 'preprune':       whether to prune the edges first 
                           (default = false)
       - 'prunemethod':    the method used to prune the edge set
                           (default = [], means using default method)
                           refer to slpruneedgeset for the specification 
                           of the prune methods.
       - 'sym':            whether to create symmetric graph
       - 'symmethod':      the method to symmetrize the graph
                           (default = [], means using default method)
                           refer to slsymedgeset for the specification.

 $ History $
   - Created by Dahua Lin, on Sep 10, 2006

CROSS-REFERENCE INFORMATION

This function calls:

• sladjlist2edgeset SLADJLIST2EDGESET Converts the adjacency list to edge set
• sledges2adjmat SLEDGES2ADJMAT Creates an adjacency matrix from edge set
• slgraphinfo SLGRAPHINFO Extracts basic information of a given graph representation
• slsymgraph SLSYMGRAPH Forces symmetry of the adjacency matrix of a graph
• slparseprops SLPARSEPROPS Parses input parameters

This function is called by:

• slgembed SLGEMBED Solves the general graph-based embedding
• sllemap SLLEMAP Solves Laplacian Eigenmap Embedding
• sllle SLLLE Performs Locally Linear Embedding
• sllle_wg SLLLE_WG Solves the Locally Linear Embedding from weight graph
• sllocaltancoords SLLOCALTANCOORDS Computes the local tangent coordinates
• sllocaltanspace SLLOCALTANSPACE Solves the local tangent spaces
• slltsa SLLTSA Performs Local Tangent Space Alignment Learning
• slnbreconweights SLNBRECONWEIGHTS Solve the optimal reconstruction weights on given neighbors
• slgbfe SLGBFE Performs Graph-based Feature Extraction Learning

SUBFUNCTIONS

• function A = change_adjmat(A0, opts)

SOURCE CODE

0001 function A = sladjmat(G, varargin)
0002 %SLADJMAT Constructs the adjacency matrix representation of a graph
0003 %
0004 % $ Syntax $
0005 %   - A = sladjmat(G, ...)
0006 %
0007 % $ Arguments $
0008 %   - G:        The input graph
0009 %   - A:        The adjacency matrix representation of the graph
0010 %
0011 % $ Description $
0012 %   - A = sladjmat(G, ...) constructs the adjacency matrix representation
0013 %     of a graph. You can specify the following properties to control
0014 %     the construction.
0015 %       - 'valtype':        the value type of the target matrix
0016 %                           - 'auto': if has value, make numeric matrix
0017 %                                     if no value, make logical matrix
0018 %                           - 'logical': make logical matrix always
0019 %                           - 'numeric': make numeric matrix always
0020 %                           (default = 'auto')
0021 %       - 'sparse':         whether to create a sparse matrix
0022 %                           (default = true)
0023 %       - 'preprune':       whether to prune the edges first
0024 %                           (default = false)
0025 %       - 'prunemethod':    the method used to prune the edge set
0026 %                           (default = [], means using default method)
0027 %                           refer to slpruneedgeset for the specification
0028 %                           of the prune methods.
0029 %       - 'sym':            whether to create symmetric graph
0030 %       - 'symmethod':      the method to symmetrize the graph
0031 %                           (default = [], means using default method)
0032 %                           refer to slsymedgeset for the specification.
0033 %
0034 % $ History $
0035 %   - Created by Dahua Lin, on Sep 10, 2006
0036 %
0037 
0038 %% parse the graph
0039 
0040 gi = slgraphinfo(G);
0041 
0042 
0043 
0044 %% main skeleton
0045 
0046 switch gi.form
0047     case 'edgeset'
0048         A = sledges2adjmat(gi.n, gi.nt, G.edges, varargin{:});
0049         
0050     case 'adjlist'
0051         if gi.valued
0052             sch = 3;
0053         else
0054             sch = 0;
0055         end
0056         edges = sladjlist2edgeset(G.targets, sch);
0057         A = sledges2adjmat(gi.n, gi.nt, edges, varargin{:});
0058         
0059     case 'adjmat'
0060         opts = struct(...
0061             'valtype', 'auto', ...
0062             'sparse', true, ...
0063             'preprune', false, ...
0064             'prunemethod', [], ...
0065             'sym', false, ...
0066             'symmethod', []);
0067         opts = slparseprops(opts, varargin{:});
0068         A = change_adjmat(G, opts);
0069 end
0070    
0071 
0072 %% auxiliary functions
0073 
0074 function A = change_adjmat(A0, opts)
0075 
0076 switch opts.valtype
0077     case 'auto'
0078         is_logic = islogical(A0);
0079     case 'logical';
0080         is_logic = true;
0081     case 'numeric'
0082         is_logic = false;
0083 end
0084 
0085 if is_logic
0086     if islogical(A0)
0087         A = A0;
0088     else
0089         A = (A0 ~= 0);
0090     end
0091 else
0092     if isa(A0, 'double')
0093         A = A0;
0094     else
0095         A = double(A0);
0096     end
0097 end
0098 
0099 if opts.sparse
0100     if ~issparse(A)
0101         A = sparse(A);
0102     end
0103 else
0104     if issparse(A)
0105         A = full(A);
0106     end
0107 end
0108 
0109 if opts.sym
0110     A = slsymgraph(A, opts.symmethod);
0111 end
0112 
0113 
0114         
0115         
0116     
0117 
0118         
0119 
0120 
0121 
0122 
0123 
0124 
0125

---

Generated on Wed 20-Sep-2006 12:43:11 by m2html © 2003