hanlp2.0代码块

Posted 2020-12-12 like1tree

import hanlp

tokenizer = hanlp.load(‘PKU_NAME_MERGED_SIX_MONTHS_CONVSEG‘)
tagger = hanlp.load(hanlp.pretrained.pos.CTB5_POS_RNN_FASTTEXT_ZH)
syntactic_parser = hanlp.load(hanlp.pretrained.dep.CTB7_BIAFFINE_DEP_ZH)
semantic_parser = hanlp.load(hanlp.pretrained.sdp.SEMEVAL16_NEWS_BIAFFINE_ZH)
print(semantic_parser([(‘蜡烛‘, ‘NN‘), (‘两‘, ‘CD‘), (‘头‘, ‘NN‘), (‘烧‘, ‘VV‘)]))

pipeline = hanlp.pipeline()     .append(hanlp.utils.rules.split_sentence, output_key=‘sentences‘)     .append(tokenizer, output_key=‘tokens‘)     .append(tagger, output_key=‘part_of_speech_tags‘)     .append(syntactic_parser, input_key=(‘tokens‘, ‘part_of_speech_tags‘), output_key=‘syntactic_dependencies‘)     .append(semantic_parser, input_key=(‘tokens‘, ‘part_of_speech_tags‘), output_key=‘semantic_dependencies‘)


text=‘‘‘
HanLP是一系列模型与算法组成的自然语言处理工具包，目标是普及自然语言处理在生产环境中的应用。
HanLP具备功能完善、性能高效、架构清晰、语料时新、可自定义的特点。
内部算法经过工业界和学术界考验，配套书籍《自然语言处理入门》已经出版。
‘‘‘
p1=pipeline(text)
type(p1)  # hanlp.common.document.Document


import pickle

pickle.dump( p1, open( "save.p1", "wb" ) )

p2 = pickle.load( open( "save.p1", "rb" ) )


with open(‘/home/chencheng/data/sogou_phone0/sogou_q_phone‘,‘r‘)as ifile:
    corpus = ifile.read()