fixing this error?

i want to calculate precision , recall , f-score and ...., but i get this erro:

Undefined function 'diag' for input arguments of type 'mlearnlib.graphics.chart.ConfusionMatrixChart'.

Error in SHIVANaugmented (line 74)
tp_m = diag(cm);
==============================================the following is a part of my code:


cm=confusionchart (imdsValidation.Labels, YPred);
%%
tp_m = diag(cm); 

 for i = 1:2 % number of classes 
    TP = tp_m(i);
    FP = sum(cm(:, i), 1) - TP;
    FN = sum(cm(i, :), 2) - TP;
    TN = sum(cm(:)) - TP - FP - FN;

    Accuracy = (TP+TN)./(TP+FP+TN+FN);

    TPR = TP./(TP + FN);%tp/actual positive  RECALL SENSITIVITY
    if isnan(TPR)
        TPR = 0;
    end
    PPV = TP./ (TP + FP); % tp / predicted positive PRECISION
    if isnan(PPV)
        PPV = 0;
    end
    TNR = TN./ (TN+FP); %tn/ actual negative  SPECIFICITY
    if isnan(TNR)
        TNR = 0;
    end
    FPR = FP./ (TN+FP);
    if isnan(FPR)
        FPR = 0;
    end
    FScore = (2*(PPV * TPR)) / (PPV+TPR);

    if isnan(FScore)
        FScore = 0;
    end
end

%%
save youRnetwork net

[Walter Roberson]

A confusion chart is a graphic object. What are you hoping that diag() of one would return?

[shivan artosh]

Number

Answer 1

[Walter Roberson]

tp_m = diag(cm.NormalizedValues); 

[shivan artosh]

excuse me i get this error:

Error using sum
Invalid data type. First argument must be numeric or logical.

Error in SHIVANaugmented (line 78)
    FP = sum(cm(:, i), 1) - TP;

[Walter Roberson]

Well, Don't Do That.

The return value from confusionchart is not a confusion matrix !!

Have you looked at https://www.mathworks.com/help/stats/confusionmat.html ?

[shivan artosh]

yes i read it but i can not fix my error

[Walter Roberson]

Go back to your original code and change

cm=confusionchart (imdsValidation.Labels, YPred);

to

cc = confusionchart (imdsValidation.Labels, YPred);
cm = cc.NormalizedValues;

[shivan artosh]

excellent it is working now. thank you