使用NLTK标记数据框

如何解决使用NLTK标记数据框

我试图标记化数据框中列的文本输入，但是我不确定所使用的函数的输入参数应该是什么。功能参数和说明如下：

def Tokenize(text,nwords):
    """
    Given list of text to be processed and a list
    of kNown words,return a list of edited and
    tokenized words.
    Spell-checking is implemented using a
    Bayesian spell-checking algorithm
    (http://norvig.com/spell-correct.html).
    By default,this is based on the Project Gutenberg
    corpus,a collection of approximately 1 million texts
    (http://www.gutenberg.org). A copy of this is included
    within this package. If desired,users may specify a
    different spell-check training corpus in the
    `training_dictionary` argument of the
    `prepare_transcripts()` function.

我应该将数据框中的列转换为列表，然后尝试将其作为文本输入吗？但是nwords参数呢？该功能需要什么样的输入？