How sequentialfs select XTRAIN and XTEST for calling fun to perform for 10-fold cross-validation?

I want to select features using sequentialfs where I used 10 for cross-validation for cvpartition. I have given the Data which is my training data (observation with feature values) and Labels which are the class labels for each observation. My code looks like the following:

Data=read from the excel file;

Labels=read from the excel file;

c=cvpartition(Labels,'KFold',10);

classf=@(Train_data, Train_labels, Test_data, Test_labels)...

sum(predict(fitcsvm(Train_data, Train_labels, 'KernenFunction', 'rbf'), Test_data) ~= Test_labels);

No_features=20;

[fs, history]=sequentialfs(classf,Data,Labels,'cv',c,'nfeatures',No_features,'direction', 'forward', 'options',opts)

But, I could not understand how the sequentialfs select the Train_data subset and Test_data subset for each fold.

It is possilbe to custom select the training and test subset for each fold.

Thanks.