神经网络 向量化的典型手法 $ from RIBES import RIBESevaluator $ ribes = RIBESevaluator() $ score = ribes.eval([target], [[references]]) $ from sumeval.metrics.rouge import RougeCalculator $ rouge = RougeCalculator(lang=‘<填en/ja/zh>’) $ Rouge_L score = rouge.rouge_l(target, references) 机器翻译的手法 $ from gensim.models import word2vec $ model = word2vec.Word2Vec(input_data, size=<..>, min_count=<..>, window=<..>) $ from sklearn.feature_extraction.text import CountVectorizer $ from sklearn.feature_extraction.text import TfidfVectorizer 为什么不用 doc2vec呢?