Recently, I want to construct a som model using somtoolbox in matlab. I have 350 SST (sea surface temperature) images. Each image is 400 pixels by 480 pixels. And each SST image was transformed into a single row vector by concatenating image rows. So that the input matrix consisted of 350 rows (image) and 192000 columns (pixels). But when I run the following code,some errors occured. Could you please tell me how to solve this problem? Thanks for your consideration.
`clc;clear;
%prepare your data (satellite remote sensing data)
myFiles=dir('D:/MatlabTemp/SSC/.tif');
fileLength=length(myFiles);
myMatrix=[];
firstRaster=imread('D:/MatlabTemp/SSC/m20030530.tif');
[m,n]=size(firstRaster);
for i=1:(fileLength)
filename=myFiles(i).name;
path='D:\MatlabTemp\SSC';
imgPath=strcat(path,'\',filename);
raster=imread(imgPath);
vector=reshape(raster,1,mn);
myMatrix=[myMatrix;vector];
end
%TSM data n * m (n samples(image)) m (pixels)
myMatrix(isnan(myMatrix)) = -1;
%inialize parameters of som algorithm
msize = [3 4];
init = 'lin';
train = 'batch';
neigh = 'ep';
rad = [5 1];
y = myMatrix;
%construct the input data used for som algorithm
sD = som_data_struct(y);
if strcmp(init,'rand')
sM = som_randinit(sD,'msize',msize);
elseif strcmp(init,'lin')
sM = som_lininit(sD,'msize',msize,'lattice','rect','shape','sheet');
end
%train som algorithm
if strcmp(train,'batch')
sM = som_batchtrain(sM, sD,'radius',rad,'trainlen',1000,'alpha_type','inv','neigh',neigh);
elseif strcmp(train,'seq')
sM = som_seqtrain(sM, sD,'radius',rad,'trainlen',1000,'neigh',neigh);
elseif strcmp(train,'sompak')
sM = som_sompaktrain(sM,sD,'radius',rad,'trainlen',1000,'neigh',neigh,'alpha_type','inv');
end`
Error using zeros
Requested 192000×192000 （274.7GB）array exceeds maximum array preference.Creation of arrays greater than this limit may take a long time and cause MATLAB to become unresponsive. See array size limit or preference panel for more information.
Recently, I want to construct a som model using somtoolbox in matlab. I have 350 SST (sea surface temperature) images. Each image is 400 pixels by 480 pixels. And each SST image was transformed into a single row vector by concatenating image rows. So that the input matrix consisted of 350 rows (image) and 192000 columns (pixels). But when I run the following code,some errors occured. Could you please tell me how to solve this problem? Thanks for your consideration.
`clc;clear; %prepare your data (satellite remote sensing data) myFiles=dir('D:/MatlabTemp/SSC/.tif'); fileLength=length(myFiles); myMatrix=[]; firstRaster=imread('D:/MatlabTemp/SSC/m20030530.tif'); [m,n]=size(firstRaster); for i=1:(fileLength) filename=myFiles(i).name; path='D:\MatlabTemp\SSC'; imgPath=strcat(path,'\',filename); raster=imread(imgPath); vector=reshape(raster,1,mn); myMatrix=[myMatrix;vector]; end
%TSM data n * m (n samples(image)) m (pixels) myMatrix(isnan(myMatrix)) = -1;
%inialize parameters of som algorithm msize = [3 4]; init = 'lin'; train = 'batch'; neigh = 'ep'; rad = [5 1];
y = myMatrix;
%construct the input data used for som algorithm sD = som_data_struct(y);
if strcmp(init,'rand') sM = som_randinit(sD,'msize',msize); elseif strcmp(init,'lin') sM = som_lininit(sD,'msize',msize,'lattice','rect','shape','sheet'); end %train som algorithm if strcmp(train,'batch') sM = som_batchtrain(sM, sD,'radius',rad,'trainlen',1000,'alpha_type','inv','neigh',neigh); elseif strcmp(train,'seq') sM = som_seqtrain(sM, sD,'radius',rad,'trainlen',1000,'neigh',neigh); elseif strcmp(train,'sompak') sM = som_sompaktrain(sM,sD,'radius',rad,'trainlen',1000,'neigh',neigh,'alpha_type','inv'); end`
Error using zeros
Requested 192000×192000 （274.7GB）array exceeds maximum array preference.Creation of arrays greater than this limit may take a long time and cause MATLAB to become unresponsive. See array size limit or preference panel for more information.
Error in som_liminit(line 222) A = zeros(dim);