Description of slpcarecon

Home > sltoolbox > subspace > slpcarecon.m

slpcarecon

PURPOSE

SLPCARECON Reconstructs the samples in original space

SYNOPSIS

function X = slpcarecon(S, Y)

DESCRIPTION

SLPCARECON Reconstructs the samples in original space

 $ Syntax $
   - Xr = slpcarecon(S, Y)

 $ Arguments $
   - S:        the PCA model struct
   - Y:        the principal component features
   - Xr:       the reconstructed samples

 $ Description $
   - Xr = slpcarecon(S, Y) reconstructs the original samples approximately
     using the principal components Y. If the dimension of Y is less than
     the subspace dimension, the leading space dimensions will be used.

 $ History $
   - Created by Dahua Lin, on Aug 17, 2006
   - Modified by Dahua Lin, on Sep 10, 2006
       - replace sladd by sladdvec to increase efficiency

CROSS-REFERENCE INFORMATION

This function calls:

• sladdvec SLADDVEC adds a vector to columns or rows of a matrix

This function is called by:

SOURCE CODE

0001 function X = slpcarecon(S, Y)
0002 %SLPCARECON Reconstructs the samples in original space
0003 %
0004 % $ Syntax $
0005 %   - Xr = slpcarecon(S, Y)
0006 %
0007 % $ Arguments $
0008 %   - S:        the PCA model struct
0009 %   - Y:        the principal component features
0010 %   - Xr:       the reconstructed samples
0011 %
0012 % $ Description $
0013 %   - Xr = slpcarecon(S, Y) reconstructs the original samples approximately
0014 %     using the principal components Y. If the dimension of Y is less than
0015 %     the subspace dimension, the leading space dimensions will be used.
0016 %
0017 % $ History $
0018 %   - Created by Dahua Lin, on Aug 17, 2006
0019 %   - Modified by Dahua Lin, on Sep 10, 2006
0020 %       - replace sladd by sladdvec to increase efficiency
0021 %
0022 
0023 %% parse and verify input
0024 
0025 if ~isstruct(S)
0026     error('sltoolbox:invalidarg', ...
0027         'S should be a PCA model struct');
0028 end
0029 
0030 if ~isnumeric(Y) || ndims(Y) ~= 2
0031     error('sltoolbox:invalidarg', ...
0032         'The features Y should be a 2D numeric matrix');
0033 end
0034 
0035 dy = size(Y, 1);
0036 if dy > S.feadim
0037     error('sltoolbox:sizmismatch', ...
0038         'The feature dimension of Y exceeds the subspace dimension preserved in model');
0039 end
0040 
0041 %% reconstruct
0042 
0043 if dy == S.feadim
0044     X = S.P * Y;
0045 else
0046     X = S.P(:, 1:dy) * Y;
0047 end
0048 
0049 X = sladdvec(X, S.vmean, 1);
0050 
0051 
0052 
0053

---

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