from sklearn.preprocessing import OneHotEncoder

cols_order = [selected_word + "_w" for selected_word in selected_words] + \