transformer/transformer.py

from sklearn.feature_extraction.text import TfidfVectorizer

def create_tfidf(dataset, feature, label):
  tfidf = TfidfVectorizer()
  return tfidf.fit(dataset[feature].astype(label))