Update text_tools.py

This commit is contained in:
yongzhuo 2019-04-29 21:55:02 +08:00 committed by GitHub
parent f56d0a182c
commit 7c10ea984b
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23

View File

@ -310,6 +310,16 @@ def judge_translate_english(sen_org, sen_tra):
return True
def load_word2vec_model(model_path, binary_type=True, encoding_type = 'utf-8', limit_words=None):
'''
下载词向量
:param model_path: str
:return: word2vec model
'''
word2vec_model = gensim.models.KeyedVectors.load_word2vec_format(model_path, binary=binary_type, limit=limit_words, encoding=encoding_type, unicode_errors='ignore')
return word2vec_model
#todo #句子改写,同义词替换,去停用词等
@ -343,4 +353,4 @@ if __name__ == '__main__':
#',0.0,1.0'
# np.savetxt('001', [word2vec_model.wv["的"], word2vec_model.wv["的"]])
# gg = np.loadtxt('001')
# gg = np.loadtxt('001')