How many test samples were allocated for each subject in the stratified train-test split (30% test size)?

How many test samples were allocated for each subject in the stratified train-test split (30% test size)?: a task in SmolDataEnvs-harbor-train-sandbox: 📊 SmolDataEnvs: Harbor (train) (Harbor dataset). Files (in /home/user/input, no subfolders): - olivetti faces.npy - olivetti faces target.npy