def pinyin_2_hanzi(self, pinyinList):
from Pinyin2Hanzi import DefaultDagParams
dagParams = DefaultDagParams()
result = None
for pinyin in pinyinList:
hanzi = [l[0] for l in dagParams.get_phrase([pinyin], 10000)]
if result is None:
result = hanzi
else:
result = [a + b for a in result for b in hanzi]
return result
def genarate_word_error(sents):
ans = []
# hmmparams = DefaultHmmParams()
dagparams = DefaultDagParams()
for sent in sents:
seg_sent = list(jieba.cut(sent))
while True:
select_word = random.sample(seg_sent, 1)[0] #随机一个候选词
if len(select_word) > 1:
break
error_word = select_word
pinyin_list = lazy_pinyin(select_word) #获取选定词的拼音
# result1 = viterbi(hmm_params=hmmparams, observations=pinyin_list, path_num=5)
try:
result2 = dag(dagparams, pinyin_list, path_num=5, log=True)
except KeyError:
continue
while len(result2) > 1:
error_word = ''.join(random.choice(result2).path)
if error_word != select_word:
break
word_index = sent.find(select_word) #替换词语中的单字
err_sent = sent[:word_index] + error_word
if word_index + len(select_word) < len(sent):
err_sent += sent[word_index + len(select_word):]
if err_sent != sent:
ans.append((sent, err_sent))
return ans
def __init__(self, w, h, x, y, font=None, callback=None):
"""
:param w:文本框宽度
:param h:文本框高度
:param x:文本框坐标
:param y:文本框坐标
:param font:文本框中使用的字体
:param callback:在文本框按下回车键之后的回调函数
"""
self.width = w
self.height = h
self.x = x
self.y = y
self.text = "" # 文本框内容
self.callback = callback
# 创建背景surface
self.__surface = pygame.Surface((w, h))
# 如果font为None,那么效果可能不太好，建议传入font，更好调节
if font is None:
self.font = pygame.font.SysFont('microsoftyaheimicrosoftyaheiui',
16)
else:
self.font = font
self.dagparams = DefaultDagParams()
self.state = 0 # 0初始状态 1输入拼音状态
self.page = 1 # 第几页
self.limit = 5 # 显示几个汉字
self.pinyin = ''
self.word_list = [] # 候选词列表
self.word_list_surf = None # 候选词surface
self.buffer_text = '' # 联想缓冲区字符串
def pinyin_2_hanzi(pinyinList):
from Pinyin2Hanzi import DefaultDagParams
from Pinyin2Hanzi import dag
dagParams = DefaultDagParams()
# 取第一个值
result = dag(dagParams, pinyinList, path_num=10, log=True)[0].path[0]
return result
def pinyin_2_hanzi(pinyinList):
from Pinyin2Hanzi import DefaultDagParams
from Pinyin2Hanzi import dag
dagParams = DefaultDagParams()
result = dag(dagParams, pinyinList, path_num=10, log=True) # 10 代表后选值个数
for item in result:
socre = item.score
res = item.path # 转换结果
print(socre, ''.join(res))
def pinyin_to_chinese(self, data):
'''get the chinese from the pinyin
:param data: pinyin data
:return:
'''
dagparames = DefaultDagParams()
result = dag(dagparames, data, path_num=10, log=True)
for item in result:
print(str(item.score) + ":", item.path)
def pinyin_2_hanzi(pinyin_str):
pinyin_list = pinyin_str.split()
dagParams = DefaultDagParams()
# 1个候选值
result = dag(dagParams, pinyin_list, path_num=1, log=True)
if result:
res = result[0].path # 转换结果
hanzi_str = ''.join(res)
return hanzi_str
else:
logger.info("转化有误：" + pinyin_str)
def pinyin_to_hanzi(pinyin, Topk=5):
'''
拼音转化为汉字
汉字存在多意性，所以这里没有一一对应的关系，只能选出概率最高的topk
'''
translator = DefaultDagParams()
result = dag(translator, pinyin, path_num=Topk, log=True)
for item in result:
socre = item.score # 得分
res = item.path # 转换结果
print socre, ''.join([one.decode('utf-8') for one in res])
def pinyin_2_hanzi(word): if Pinyin2Hanzi.is_chinese(word): word_pinyin = lazy_pinyin(word) dagParams = DefaultDagParams() word_list = [] result = dag(dagParams, word_pinyin, path_num=3, log=True) for item in result: word_list.append(item.path[0]) return word_list else: return "Null"
def pinyin_2_hanzi(sentences):
from Pinyin2Hanzi import DefaultDagParams
from Pinyin2Hanzi import dag
dagParams = DefaultDagParams()
pinyinList = lazy_pinyin(sentences)
print(pinyinList)
result = dag(dagParams, pinyinList, path_num=3) #10代表侯选值个数
for item in result:
socre = item.score
res = item.path # 转换结果
print(socre, res)
def pinyin_2_hanzi(self, pinyinList):
dagParams = DefaultDagParams()
result = dag(dagParams, pinyinList, path_num=10, log=True) #10代表侯选值个数
item_result = []
try:
for item in result:
# socre = item.score
# res = item.path # 转换结果
item_result.append([item.score, item.path])
return sorted(item_result, key=itemgetter(0), reverse=True)[0][1]
except Exception as e:
print(e)
print("输入异常，请重新输入拼音")
def pinyin2hanzi(pinyin_list):
'''
:param text_list: 拼音列表
:return: 文本二维列表，且每个一维列表的长度为1
'''
dagParams, entities = DefaultDagParams(), []
for line in pinyin_list:
result = dag(dagParams, line, path_num=5, log=True)
for item in result:
res = item.path # 转换结果
if len(res) > 1:
continue
entities.append(res)
return entities
def pinyin_to_hanzi(pinyin,Topk=5,Log=True):
'''
拼音转化为汉字
汉字存在多意性，所以这里没有一一对应的关系，只能选出概率最高的topk
'''
print(pinyin)
translator=DefaultDagParams()
result=dag(translator,pinyin,path_num=Topk,log=Log)
#print(result)
# for item in result:
# socre=item.score # 得分
# res=item.path # 转换结果
# print(socre, ''.join([one.decode('utf-8') for one in res]))
return result
def pinyin_2_hanzi(pinyin_str):
'''
zhao qing shi ding hu qu fang di chan xie hui --- 肇庆市鼎湖区房地产协会
'''
pinyin_list = pinyin_str.split()
dagParams = DefaultDagParams()
# 1个候选值
result = dag(dagParams, pinyin_list, path_num=1, log=True)
if result:
res = result[0].path # 转换结果
hanzi_str = ''.join(res)
return hanzi_str
else:
return ''
logger.info("转化有误：" + pinyin_str)
def __init__(self):
# 初始化
self.hmmparams = DefaultHmmParams()
self.dagparams = DefaultDagParams()
self.result = ''
self.shengmu = ['b', 'c', 'd', 'f', 'g', 'h', 'j', 'k', 'l', 'm', 'n', 'p', 'q', 'r', 's', 't', 'w', 'x', 'y',
'z', 'ch', 'sh', 'zh']
self.yy = ['a', 'ai', 'an', 'ang', 'ao', 'e', 'en', 'eng', 'er', 'o', 'ou', 'ong']
self.ym_b = ["a", "ai", "an", "ang", "ao", "ei", "en", "eng", "i", "ian", "iao", "ie", "in", "ing", "o", "u"]
self.ym_c = ["a", "ai", "an", "ang", "ao", "e", "en", "eng", "i", "ong", "ou", "u", "uan", "ui", "un", "uo"]
self.ym_d = ["a", "ai", "an", "ang", "ao", "e", "ei", "en", "eng", "i", "ia", "ian", "iao", "ie", "ing", "iu",
"ong", "ou", "u", "uan", "ui", "un", "uo"]
self.ym_f = ["a", "an", "ang", "ei", "en", "eng", "iao", "o", "ou", "u"]
self.ym_g = ["a", "ai", "an", "ang", "ao", "e", "ei", "en", "eng", "ong", "ou", "u", "uai", "uan", "uang", "ui",
"un", "uo"]
self.ym_h = ["a", "ai", "an", "ang", "ao", "e", "ei", "en", "eng", "ong", "ou", "u", "ua", "uai", "uan", "uang",
"ui", "un", "uo"]
self.ym_j = ["i", "ia", "ian", "iang", "iao", "ie", "in", "ing", "iong", "iu", "u", "uan", "ue", "un"]
self.ym_k = ["a", "ai", "an", "ang", "ao", "e", "en", "eng", "ong", "ou", "u", "ui", "un", "uo"]
self.ym_l = ["a", "ai", "an", "ang", "ao", "e", "ei", "eng", "i", "ia", "ian", "iao", "ie", "in", "ing", "iu",
"o", "ong", "ou", "u", "uan", "un", "uo", "v", "ve"]
self.ym_m = ["a", "ai", "an", "ang", "ao", "e", "ei", "en", "eng", "i", "ian", "iao", "ie", "in", "ing", "iu",
"o", "ou", "u"]
self.ym_n = ["a", "ai", "an", "ang", "ao", "e", "ei", "en", "eng", "i", "ian", "iang", "iao", "ie", "in", "ing",
"iu", "ong", "ou", "u", "uan", "un", "uo", "v", "ve"]
self.ym_p = ["a", "ai", "an", "ang", "ao", "e", "ei", "en", "eng", "i", "ian", "iao", "ie", "in", "ing", "o",
"ou", "u"]
self.ym_q = ["i", "ia", "ian", "iang", "iao", "ie", "in", "ing", "iong", "iu", "u", "uan", "ue", "un"]
self.ym_r = ["an", "ang", "ao", "e", "en", "eng", "i", "ong", "ou", "u", "ua", "uan", "ui", "un", "uo"]
self.ym_s = ["a", "ai", "an", "ang", "ao", "e", "en", "eng", "i", "ong", "ou", "u", "uan", "ui", "un", "uo"]
self.ym_t = ["a", "ai", "an", "ang", "ao", "e", "ei", "eng", "i", "ian", "iao", "ie", "ing", "ong", "ou", "u",
"uan", "ui", "un", "uo"]
self.ym_w = ["a", "ai", "an", "ang", "ei", "en", "eng", "o", "u"]
self.ym_x = ["i", "ia", "ian", "iang", "iao", "ie", "in", "ing", "iong", "iu", "u", "uan", "ue", "un"]
self.ym_y = ["a", "an", "ang", "ao", "e", "i", "in", "ing", "o", "ong", "ou", "u", "uan", "ue", "un"]
self.ym_z = ["a", "ai", "an", "ang", "ao", "e", "ei", "en", "eng", "i", "ong", "ou", "u", "uan", "ui", "un",
"uo"]
self.ym_ch = ["a", "ai", "an", "ang", "ao", "e", "en", "eng", "i", "ong", "ou", "u", "ua", "uai", "uan", "uang",
"ui", "un", "uo"]
self.ym_sh = ["a", "ai", "an", "ang", "ao", "e", "ei", "en", "eng", "i", "ou", "u", "ua", "uai", "uan", "uang",
"ui", "un", "uo"]
self.ym_zh = ["a", "ai", "an", "ang", "ao", "e", "ei", "en", "eng", "i", "ong", "ou", "u", "ua", "uai", "uan",
"uang", "ui", "un", "uo"]
self.ym = [self.yy, self.ym_b, self.ym_c, self.ym_d, self.ym_f, self.ym_g, self.ym_h, self.ym_j, self.ym_k,
self.ym_l, self.ym_m, self.ym_n, self.ym_p, self.ym_q, self.ym_r, self.ym_s, self.ym_t, self.ym_w,
self.ym_x, self.ym_y, self.ym_z, self.ym_ch, self.ym_sh, self.ym_zh
]
def get_chengyu(word):
bigger_list = []
if Pinyin2Hanzi.is_chinese(word):
word_pinyin = lazy_pinyin(word)
dagParams = DefaultDagParams()
word_list = []
result = dag(dagParams, word_pinyin, path_num=3, log=True)
for item in result:
word_list.append(item.path[0])
for avg_word in word_list:
bigger_list.append(find_chengyu(avg_word))
heiheihei = list((chain(*bigger_list)))
max_len = len(heiheihei) - 1
flag = random.randint(0, max_len)
return heiheihei[flag]
else:
return "Null"
def __init__(self, w, h, x, y, callback=None):
self.font = pygame.font.SysFont('microsoftyaheimicrosoftyaheiui', 16)
self.width = w
self.height = h
self.x = x
self.y = y
self.text = "" # 文本框内容
self.callback = callback
# 创建背景surface
self.__surface = pygame.Surface((w, h))
self.__surface.fill((250, 250, 250))
self.dagparams = DefaultDagParams()
self.state = 0 # 0初始状态 1输入拼音状态
self.page = 1 # 第几页
self.limit = 5 # 显示几个汉字
self.pinyin = ''
self.word_list = [] # 候选词列表
self.word_list_surf = None # 候选词surface
self.buffer_text = '' # 联想缓冲区字符串
# coding: utf-8
from __future__ import (print_function, unicode_literals)
import sys
sys.path.append('..')
from Pinyin2Hanzi import DefaultDagParams
from Pinyin2Hanzi import dag
dagparams = DefaultDagParams()
result = dag(dagparams, ['wo'])
for item in result:
print(item.score, '/'.join(item.path))
print(20 * '*')
result = dag(dagparams, ['ni', 'hao'])
for item in result:
print(item.score, '/'.join(item.path))
print(20 * '*')
result = dag(dagparams, ['ni', 'bu', 'zhi', 'dao', 'de', 'shi'])
for item in result:
print(item.score, '/'.join(item.path))
print(20 * '*')
result = dag(dagparams, ['ni', 'bu', 'zhi', 'dao', 'de', 'shi'],
path_num=2,
def __init__(self):
self._dagparams = DefaultDagParams()
#!/usr/bin/env python
# -*- coding:utf-8 -*-
# @FileName :input.py
# @Time :2022/1/4 21:04
# @Author :russionbear
import pygame
from Pinyin2Hanzi import is_pinyin, dag, DefaultDagParams
from .. import Pen
import ctypes
__param = DefaultDagParams()
PATH_NUM = 20
def get_pinyin(s0):
"""
将一段拼音，分解成一个个拼音
:param s0: 匹配的字符串
:return: 匹配到的拼音列表
"""
for i1, i in enumerate(reversed(s0)):
if i < 'a' or i > 'z':
s0 = s0[-i1 + 1:]
result = []
if not s0:
return result
max_len = 6 # 拼音最长为6
def pinyin_2_hanzi(pinyinList):
from Pinyin2Hanzi import DefaultDagParams
from Pinyin2Hanzi import dag
dagParams = DefaultDagParams()
result = dag(dagParams, pinyinList, path_num=500000, log=True)#10代表侯选值个数
return [''.join(item.path) for item in result]
# coding: utf-8
from __future__ import (print_function, unicode_literals)
import sys
sys.path.append('..')
from Pinyin2Hanzi import all_pinyin
from Pinyin2Hanzi import DefaultDagParams
dagparams = DefaultDagParams()
for py in all_pinyin():
if len(dagparams.get_phrase([py]) ) == 0:
print(py)
print( dagparams.get_phrase(['ju']) )
print( dagparams.get_phrase(['jve']) )
