Stratified K-fold validation using crossval funtion

I am trying to customize the "Weighted kNN"-based classification code generated after a 10-fold cross validation on my data using the Classification Learner App (Using the Generate Function option in the app).

Query 1:

partitionedModel = crossval(trainedClassifier.ClassificationKNN, 'KFold',10);

After generating the code, I found that the above code is responsible for partitioning the data. I have an imbalanced dataset and I believe that "Stratified" k-fold cross validation ensures equal amount of data is considered from each class.

Query: Does the above code automatically perform Stratified k-fold cross validation? If not, how do I include it in the code?

Answer 1

[Rahul]

Hi @Jhon Gray,

You correctly identified 'crossval' function to be responsible for partitioning the data. Considering you are applying 'KFold' cross validation to your data, the 'crossval' function would stratify the data by default i.e. ensure equal amount of data is considered from each class.

For other categories of cross-validation tasks based on classification or if a particular class or list of class labels need to be specified to be specifically stratified, the 'Stratify' property of the 'crossval' function needs to be adjusted apprpriately.

The following MATLAB Answer can be referred:

https://www.mathworks.com/matlabcentral/answers/271843-how-does-crossval-for-k-fold-cv-work-in-matlab-after-training-a-classifier

The following MathWorks documentations can be referred to know more:

'Stratify' (crossval): https://www.mathworks.com/help/stats/crossval.html#mw_35debc35-cab8-4ad9-a97d-b039ac03a22b

Thanks.