class my_SKF:
def __init__(self, X, y, splits):
self.X = X
self.y = y
self.skf = StratifiedKFold(n_splits=splits)
self.skf = self.skf.split(X, y)
def __next__(self):
train_index, test_index = self.skf.__next__()
X_train, X_test = self.X[train_index], self.X[test_index]
y_train, y_test = self.y[train_index], self.y[test_index]
return X_train, y_train, X_test, y_test
def __iter__(self):
return self
# # # Data Recovery # # # ################# X, y, dim = data_recovery(opt, date) #################### # # # Train Test Split # # # #################### kf = StratifiedKFold(n_splits=5, random_state=opt.seed, shuffle=True) kf = kf.split(X, y) train_index, test_index = kf.__next__() ######## # # # MLFP # # # ######## ###### # # # CV # # # ###### kf = StratifiedKFold(n_splits=opt.nb_cv, random_state=opt.seed, shuffle=True)
