如何摆脱以下错误：“ ValueError：'00'不在列表中”？

Question

我是python的新手，正在尝试解析朴素贝叶斯算法（多项式模型），但是我的代码出现了一些错误。

这是算法： 到目前为止，这是我的代码：

from collections import Counter
import nltk
from nltk.corpus import stopwords

def CountDocs(D):
    D_length = X_train_counts.shape[0] # number of documents
    return D_length

def ExtractVocabulary(D):
    vectorizer = CountVectorizer()
    X = vectorizer.fit_transform(D.data)
    return vectorizer.get_feature_names()


def CountDocsInClass(c):
    data_from_class = fetch_20newsgroups(subset='train', categories=[c]).data
    return len(data_from_class)

def ConcatenateTextOfAllDocsInClass(data_from_class):
    for i in range(0, len(data_from_class)):
        if i not in stop_words:
            text += word_tokenize[i]
    return text

def TrainMultinomialNB(C,D):
    V = D.data
    N = CountDocs(D)
    ClassLength = {}
    prior = {}
    condprob = {}

    for c in C:
        class_length = CountDocsInClass(c)
        print(data_from_class)
        ClassLength.update({c: nc})
        prior.update({c:class_length/N})

        textC = ConcatenateTextOfAllDocsInClass(c)

        T_ct = Counter(textC)
        T_ct_sum = sum(T_ct[t] + 1 for t in V)

        for t in V:
            if t not in condprob:
                condprob.update({t: {c: (T_ct[t] + 1)/T_ct_sum}})
            else:
                condprob[t].update({c: (T_ct[t] + 1)/T_ct_sum})

    return V, prior, condprob

TrainMultinomialNB(feature_names, twenty_train)

我在其中应用此算法的数据： 变量feature_names代表课程，变量twenty_train代表训练集。 x_train_counts是上面这段代码的使用，它是：X_train_counts = count_vect.fit_transform(twenty_train.data) data_from_class变量加载20个Newsgrous数据集-一组12k个文档的集合，分为c个不同类别

错误：