How to generate unique score using vectorizer.transform(x_train)

Question

I am trying to generate unique scores for words with tfidf vectorizer, but unable to generate that. Following is the code-

def prepare_data(preprocessed,label