Main Content

resubMargin

Resubstitution classification margins for discriminant analysis classifier

collapse all in page

    Syntax

    m = resubMargin(Mdl)

    Description

    m = resubMargin(Mdl) returns the resubstitution classification margins values m for the trained discriminant analysis classifier Mdl using the training data stored in Mdl.X and the corresponding true class labels stored in Mdl.Y. m is returned as a numeric column vector with the same length as Mdl.Y. The software estimates each entry of m using the trained discriminant analysis classifier Mdl, the corresponding row of Mdl.X, and the true class label Mdl.Y.

    example

    Examples

    collapse all

    Open Live Script

    Find the margins for a discriminant analysis classifier for Fisher's iris data by resubstitution. Examine several entries.

    Load Fisher's iris data set.

    load fisheriris

    Train a discriminant analysis classifier.

    Mdl = fitcdiscr(meas,species);

    Compute the resubstitution margins, and display several of them.

    m = resubMargin(Mdl);
m(1:25:end)
    ans = 6×1

    1.0000
    1.0000
    0.9998
    0.9998
    1.0000
    0.9946

    Input Arguments

    collapse all

    Discriminant analysis classifier, specified as a ClassificationDiscriminant model object trained with fitcdiscr.

    More About

    collapse all

    Classification Margin

    The classification margin is the difference between the classification score for the true class and maximal classification score for the false classes.

    The classification margin is a column vector with the same number of rows as in the matrix X. A high value of margin indicates a more reliable prediction than a low value.

    Score

    For discriminant analysis, the score of a classification is the posterior probability of the classification. For the definition of posterior probability in discriminant analysis, see Posterior Probability.

    Version History

    Introduced in R2011b

    expand all

    See Also

    Classes

    Functions

    Topics