Python Gensim在语料库中提取单词的TF-IDF值

Python Gensim在语料库中提取单词的TF-IDF值,python,text,tf-idf,gensim,Python,Text,Tf Idf,Gensim,我使用的是Gensim TFIDF模型。这是我的代码: dictionary = corpora.Dictionary(line.lower().split()) for line in open('aaa.txt')) class MyCorpus(object): def __iter__(self): for line in open('aaa.txt'): yield dictionary.doc2bow(line.lower().spl

我使用的是Gensim TFIDF模型。这是我的代码:

dictionary = corpora.Dictionary(line.lower().split()) for line in open('aaa.txt'))

class MyCorpus(object):
    def __iter__(self):
        for line in open('aaa.txt'):
            yield dictionary.doc2bow(line.lower().split())

corpus = MyCorpus()

tfidf = models.TfidfModel(corpus)

corpus_tfidf = tfidf[corpus]
现在我想提取每个单词的tf-idf值,我知道它们在corpus_-tfidf变量中,我尝试了下面的代码来查看所有单词tf-idf,但我有一个像“香蕉”这样的单词,我想找到它的tf-idf值。在字典中可以找到每个单词,比如dictionary.token2id['banana'],但是如何获得每个单词的tf-idf呢

{dictionary.get(id): value for doc in corpus_tfidf for id, value in doc}
我的语料库有6501598个文档,585499个特征,64106768个非零条目,在最短的时间内获得每个单词的值很重要