9cabb26008ca466ff5cc9f1651162757917a8e9a,skater/core/local_interpretation/text_interpreter.py,,vectorize_as_tf_idf,#Any#,45

Before Change

def vectorize_as_tf_idf(data):
    tfidf_vec = TfidfVectorizer(sublinear_tf=True, max_df=0.5,
                    stop_words="english")
    X = tfidf_vec.fit_transform(data)
    return tfidf_vec, X

After Change

    Term Frequency times Inverse Document Frequency
    // TODO: extend support to other forms of Vectorization schemes - Feature Hashing
    // Converting raw document to tf-idf feature matrix
    tfidf_vec = TfidfVectorizer(sublinear_tf=kwargs["sublinear_tf"], max_df=kwargs["max_df"],
                    stop_words=kwargs["stop_words"], smooth_idf=kwargs["smooth_idf"],
                                ngram_range=kwargs["ngram_range"])
    X = tfidf_vec.fit_transform(data)

Instances