POS标签上的word2vec

Question

请检查我获取POS矢量的代码。而不是获取POS标签矢量我只是在POS.Eg中获取字母矢量而不是获取POS标签矢量CC，DT，PRP等我得到C，D和矢量p上。

#get word and pos tagger
def get_pos_tagger(self, document):

    # tokenizer
    tokens = nltk.word_tokenize(document)
    # get pos tagger
    posTagger = nltk.pos_tag(tokens=tokens)
    tags = []
    for (word, tag) in posTagger:
        tags.append(tag)

    return tags


def get_tag_and_training_data(self):
    tags=[]
    documents=[]
    line_counter=1

    with open(self.filename) as csvfile:
        spamreader = csv.reader(csvfile, delimiter=",")
        for line in spamreader:
            #Initialize the token list for line
            tags.append(int(line[0]))
            documents.append(line[1].lower() + " " + line[2].lower())

    return tags,documents


# build pos model
def buildPosModel(self):
    tags, documents = self.get_tag_and_training_data()
    sentences = []

    for document in documents:
     sentences += self.get_pos_tagger(document)
     print(sentences)
    modelPos = gensim.models.Word2Vec(sentences=sentences, size=100, min_count=1, window=5, workers=cores)
    modelPos.wv.save_word2vec_format('word2vecposmodel.bin', binary=False)
    return modelPos