I have a very unbalanced dataset ,, thousands of healthy participants and 21 patients (16 male and 5 female) ,, I want to use bootstrapping to define a new sampler but with
