Documentation

This is machine translation

Translated by Microsoft
Mouse over text to see original. Click the button below to return to the English verison of the page.

confusion

Classification confusion matrix

Syntax

[c,cm,ind,per] = confusion(targets,outputs)

Description

[c,cm,ind,per] = confusion(targets,outputs) takes these values:

targets

S-by-Q matrix, where each column vector contains a single 1 value, with all other elements 0. The index of the 1 indicates which of S categories that vector represents.

outputs

S-by-Q matrix, where each column contains values in the range [0,1]. The index of the largest element in the column indicates which of S categories that vector represents.

and returns these values:

c

Confusion value = fraction of samples misclassified

cm

S-by-S confusion matrix, where cm(i,j) is the number of samples whose target is the ith class that was classified as j

ind

S-by-S cell array, where ind{i,j} contains the indices of samples with the ith target class, but jth output class

per

S-by-4 matrix, where each row summarizes four percentages associated with the ith class:

per(i,1) false negative rate
          = (false negatives)/(all output negatives)
per(i,2) false positive rate
          = (false positives)/(all output positives)
per(i,3) true positive rate
          = (true positives)/(all output positives)
per(i,4) true negative rate
          = (true negatives)/(all output negatives)

[c,cm,ind,per] = confusion(TARGETS,OUTPUTS) takes these values:

targets

1-by-Q vector of 1/0 values representing membership

outputs

S-by-Q matrix, of value in [0,1] interval, where values greater than or equal to 0.5 indicate class membership

and returns these values:

c

Confusion value = fraction of samples misclassified

cm

2-by-2 confusion matrix

ind

2-by-2 cell array, where ind{i,j} contains the indices of samples whose target is 1 versus 0, and whose output was greater than or equal to 0.5 versus less than 0.5

per

2-by-4 matrix where each ith row represents the percentage of false negatives, false positives, true positives, and true negatives for the class and out-of-class

Examples

[x,t] = simpleclass_dataset;
net = patternnet(10);
net = train(net,x,t);
y = net(x);
[c,cm,ind,per] = confusion(t,y)

See Also

|

Introduced in R2008a

Was this topic helpful?