def correct(sentence):
"""
句子改错
:param sentence: 句子文本
:return: 改正后的句子, list(wrongs, rights, begin_idx, end_idx)
"""
detail = []
maybe_error_ids = get_sub_array(detect(sentence))
# print('maybe_error_ids:', maybe_error_ids)
# 取得字词对应表
index_char_dict = dict()
for index in maybe_error_ids:
if len(index) == 1:
# 字
index_char_dict[','.join(map(str, index))] = sentence[index[0]]
else:
# 词
index_char_dict[','.join(map(
str, index))] = sentence[index[0]:index[-1]]
for index, item in index_char_dict.items():
# 字词纠错
sentence, detail_word = _correct_item(sentence, index, item)
if detail_word:
detail.append(detail_word)
return sentence, detail
def correct(sentence, param_ec = 1.5, param_gd = 2.5):
detail = []
# # detecting for errors
maybe_error_ids = get_valid_sub_array(sentence, get_sub_array(detect(sentence)))
# # transfer index of error chars into pairs of (idx, error_chars)
suspect_chars = [[','.join([str(i[0]), str(i[-1])]), sentence[i[0]: i[-1]]] for i in maybe_error_ids]
# # statistical correction
sentence, detail_stat = correct_stat(sentence, suspect_chars, param_ec, param_gd)
detail += detail_stat
# # rule-based correction
sentence, detail_rule = correct_rule(sentence)
detail += detail_rule
return sentence, detail
def correct(sentence):
"""
句子改错
:param sentence: 句子文本
:return: 改正后的句子, list(wrongs, rights, begin_idx, end_idx)
"""
detail = []
maybe_error_ids = get_sub_array(detect(sentence))
# print('maybe_error_ids:', maybe_error_ids)
# 取得字词对应表
index_char_dict = dict()
for index in maybe_error_ids:
if len(index) == 1:
# 字
index_char_dict[','.join(map(str, index))] = sentence[index[0]]
else:
# 词
index_char_dict[','.join(map(str, index))] = sentence[index[0]:index[-1]]
for index, item in index_char_dict.items():
# 字词纠错
sentence, detail_word = _correct_item(sentence, index, item)
if detail_word:
detail.append(detail_word)
return sentence, detail
# -*- coding: utf-8 -*- """ @author:XuMing（[email protected]) @description: """ from pycorrector import detector # fixed idx_errors = detector.detect('vd') print(idx_errors)
# -*- coding: utf-8 -*- """ @author:XuMing（[email protected]) @description: """ from pycorrector import detector idx_errors = detector.detect('少先队员因该为老人让坐') print(idx_errors)