crossval

Cross-validated discriminant analysis classifier

Syntax

cvmodel= crossval(obj) cvmodel= crossval(obj,Name,Value)

Description

cvmodel= crossval(obj)creates a partitioned model fromobj, a fitted discriminant analysis classifier. By default,crossvaluses 10-fold cross validation on the training data to createcvmodel.

cvmodel= crossval(obj,Name,Value)creates a partitioned model with additional options specified by one or moreName,Valuepair arguments.

Input Arguments

obj

Discriminant analysis classifier, produced usingfitcdiscr.

Name-Value Arguments

Specify optional pairs of arguments asName1=Value1,...,NameN=ValueN, whereNameis the argument name andValueis the corresponding value. Name-value arguments must appear after other arguments, but the order of the pairs does not matter.

Before R2021a, use commas to separate each name and value, and encloseNamein quotes.

`CVPartition`	Object of class`cvpartition`, created by the`cvpartition`function.`crossval`splits the data into subsets with`cvpartition`. Use only one of these options at a time:`'CVPartition'`,`'Holdout'`,`'KFold'`, or`'Leaveout'`. Default:`[]`
`Holdout`	Holdout validation tests the specified fraction of the data, and uses the rest of the data for training. Specify a numeric scalar from`0`to`1`. Use only one of these options at a time:`'CVPartition'`,`'Holdout'`,`'KFold'`, or`'Leaveout'`.
`KFold`	Number of folds to use in a cross-validated classifier, a positive integer value greater than 1. Use only one of these options at a time:`'CVPartition'`,`'Holdout'`,`'KFold'`, or`'Leaveout'`. Default:10
`Leaveout`	Set to`'on'`for leave-one-out cross validation. Use only one of these options at a time:`'CVPartition'`,`'Holdout'`,`'KFold'`, or`'Leaveout'`.

Examples

Create a classification model for the Fisher iris data, and then create a cross-validation model. Evaluate the quality the model usingkfoldLoss.

load fisheriris obj = fitcdiscr(meas,species); cvmodel = crossval(obj); L = kfoldLoss(cvmodel) L = 0.0200

Tips

Assess the predictive performance ofobjon cross-validated data using the “kfold” methods and properties ofcvmodel, such askfoldLoss.

Alternatives

You can create a cross-validation classifier directly from the data, instead of creating a discriminant analysis classifier followed by a cross-validation classifier. To do so, include one of these options infitcdiscr:'CrossVal','CVPartition','Holdout','KFold', or'Leaveout'.