How to do under-sampling with K-fold validation in machine learning

Question

I\'m dealing with a highly imbalanced dataset for my project rn(Ex: 1:9 in target variable), and I have already used X_train, X_test,y_train,y_test = train_test_split(X, y