Why it is 94? corss-validation for regression tree problem.

Hi,

Example shown on the page below.

https://uk.mathworks.com/help/stats/regressiontree.crossval.html

load carsmall
X = [Acceleration Displacement Horsepower Weight]; % X is 100 by 4.

Mdl = fitrtree(X,MPG); % grow a regression tree using the entire data set.

CVMdl = crossval(Mdl); % cross-validate the regression tree using 10-fold cross-validation

As I looked into CVMdl (the RegressionPartitionedModel cross-validated model), it shows the observation number used in each fold is 94. Why is it 94 not 90? See screenshot below.

Since X is 100 by 4, showing there are 100 observations to start. To have 10 folds, it means they should divide X into 90 training data and 10 test data. So I couldn't figure out why it is 94 training data instead of 90.

Please help.