Undefined function 'min' for input arguments of type 'struct'.

Undefined function 'min' for input arguments of type 'struct'.

Error in mapminmax.create (line 12)
    xmin = nnet.array.safeGather(min(x,[],2));

Error in mapminmax (line 51)
[y,settings] = mapminmax.create(x,param);

Error in GLCM_MainTotalDemo (line 46)
 P_train=mapminmax(P_train,0,1);

%********* program **************************
clear all; 
clc; 

%% BUILD DORSAL HAND VEIN TEMPLATE DATABASE 

tic;                                                                       %% calculating elapsed time for execution

%% load mat files

 load('db5.mat');
 load('db6.mat');
 
%% reshape into row vector
 
reduced_testdata = reshape(reduced_testdata,1,4,10);        % one row,four column and 15(60/4) group for 20 classes
reduced_traindata = reshape(reduced_traindata,1,4,20);       % one row,four column and 45(180/4) group for 20 classes
 
%% adjust dimension
% Adjust matrix dimension
 
P_test = cell2mat(reduced_testdata);                                         % Convert cell array to matrix
 
P_train = cell2mat(reduced_traindata); 
  
%% rearranges the dimensions of P_test and P_train 

 C = permute(P_test,[1 3 2]);
 P_test = reshape(C,[],size(P_test,2),1);
 
 C = permute(P_train,[1 3 2]);
 P_train = reshape(C,[],size(P_train,2),1);
 

%% labeling class

  train_label=load('train_label_2.txt');  
  test_label=load('test_label_2.txt');   
 
% %%% Normalisation
% 
%    P_train=P_train/256;
%    P_test=P_test/256;

%% Normalisation by min max

 P_train=mapminmax(P_train,0,1);
 P_test=mapminmax(P_test,0,1);

%% %%PCA low dimension reduction

P_train = P_train';
%%% if classes are 20 then eiganvectors not exceed then 179 
model = perform_pca(P_train,rank(P_train)-1);                       %rank(P_train)-1
test_features= linear_subspace_projection(P_test, model, 1);        
P_train=model.train';
P_test=test_features';
%% classfication 
 predictlabel = knnclassify(P_test, P_train, train_label,3,'euclidean','nearest');
 cp = classperf(test_label,predictlabel);
 Conf_Mat = confusionmat(test_label,predictlabel);
 disp(Conf_Mat);

  [c_matrix,Result,RefereceResult]= confusion.getMatrix(test_label,predictlabel); 
 
 %%  % Evaluate Performance
 [FPR, TPR,Thr, AUC, OPTROCPT]  = perfcurve(predictlabel, test_label,1);
 
 figure,
 
 plot(TPR,FPR,'r-','LineWidth',1);
 xlabel('False positive rate')
 ylabel('True positive rate') 
 title('ROC Curve for Classification ')
 Tbl = table(FPR, TPR, Thr)
 
%% FAR = FPR = FP/(FP + TN)  and  FRR = FNR = FN/(FN + TP)
 
 fprintf('\n\n Overall accuracy:%f%%\n',cp.CorrectRate*100);

%% calculating elapsed time for execution
toc

[dpb]

Don't have NN TB so can't run but the above code doesn't look like is a struct passed; you sure you're looking at the right code that generated the error?

Set a breakpoint and show us what

whos P_train

returns when breaks on entry to the subject line...

[Walter Roberson]

We need the two .txt files to test with as well.

Check the result of load() of the .txt files: it is potentially returning a struct rather than numeric.

[dpb]

That's so, Walter, but the posted code doesn't use the result of those text files to get to the line that shows the error...so is still a mystery.

[Balaji M. Sontakke]

sir, I have called following function in above code I think struct is created in following code also attaching txt files..

function [stats] = GLCM_feature_vec(IMG)
% stat features
GLCM2 = graycomatrix(IMG);                                        %GLCM2 = graycomatrix(I,'Offset',[2 0;0 2]);
%GLCM2 = graycomatrix(IMG,'Offset',[2 0;0 2]);
 stats = graycoprops(GLCM2,{'contrast','homogeneity','Correlation', 'Energy'});
end

[Balaji M. Sontakke]

Finally i got solution

stats = graycoprops(GLCM2,{'contrast','homogeneity','Correlation', 'Energy'});
a = stats.Contrast;
b = stats.Correlation;
c = stats.Energy;
d = stats.Homogeneity;
M=[a    b     c   d ]';

[dpb]

I can see how that solves a problem with respect to stats struct; I don't see how it really has anything to do with the original posting/code, though...

[Walter Roberson]

The original db5.mat and db6.mat contain cell array of struct that are converted to array of struct and those are passed to mapminmax() . However, mapminmax() can only operate on numeric values. The problem was that the source data was in the wrong form.

It appears that the struct was being created by calling GLCM_feature_vec which was returning a struct. Feature vectors are never structs, though, they must always be numeric. The modified version of the code appears to extract the numeric parts acceptably.

[dpb]

Huh. I'd forgotten cell2mat will return an array of an underlying type including a struct array...was thinking that would have had to have erred if result wasn't a "real" matrix.

[Walter Roberson]

It depends on the release which datatypes are supported by cell2mat()

[dpb]

Well, mayhaps hadn't forgotten but had never realized... :)