如何在使用tfidf.vectorizer后获得每类词的最高tf-IDF值

from sklearn.feature_extraction.text import TfidfVectorizer
import numpy as np

documents = [
    "This is document 1.",
    "This is document 2.",
    "Document 3 is different from the others."
]

vectorizer = TfidfVectorizer()
tfidf_matrix = vectorizer.fit_transform(documents)

tfidf_array = tfidf_matrix.toarray()

feature_names = vectorizer.get_feature_names()
max_tfidf_per_class = []

for i in range(tfidf_array.shape[0]):
    max_tfidf_idx = np.argmax(tfidf_array[i])
    max_tfidf_value = tfidf_array[i][max_tfidf_idx]
    max_tfidf_word = feature_names[max_tfidf_idx]
    max_tfidf_per_class.append((max_tfidf_word, max_tfidf_value))

for word, value in max_tfidf_per_class:
    print("Word: {}, TF-IDF value: {}".format(word, value))