def get(self):
params = self.parser.parse_args()
sourceModelPinYinName = pinyin.get(params.get('sourceModelName'),
format='strip')
targetModelPinYinName = pinyin.get(params.get('targetModelName'),
format='strip')
os.rename('data/' + sourceModelPinYinName,
'data/' + targetModelPinYinName)
if os.path.exists("model/" + sourceModelPinYinName):
os.rename('model/' + sourceModelPinYinName,
'model/' + targetModelPinYinName)
return {"code": 1, "message": "ok", "data": targetModelPinYinName}
def image_upload():
if request.method == 'POST':
# 通过POST方式上传图片
image_file = request.files['file']
filename = secure_filename(
pinyin.get(image_file.filename, format="numerical"))
if allowed_file(filename):
image_entity = ImageEntity()
image_entity.name = image_file.filename
image_entity.type = image_file.content_type
image_entity.data = image_file.read()
image_entity.time = datetime.datetime.now()
md5 = hashlib.md5()
md5.update(
filename +
datetime.datetime.strftime(image_entity.time, '%Y%m%d%H%M%S'))
image_entity.md5_name = md5.hexdigest() + '.' + filename.rsplit(
'.', 1)[1]
db.session.add(image_entity)
db.session.commit()
return redirect('/attach/' + image_entity.md5_name)
else:
result = ResultModel(ResultModel.FAILED_CODE, '图片格式不合法', None)
return jsonify(vars(result))
return render_template('upload.html')
def get(self):
params = self.parser.parse_args()
return {
"code": 1,
"message": "ok",
"data": pinyin.get(params.get('data'), format='strip')
}
def get(self):
params = self.parser.parse_args()
modelName = params.get('modelName')
modelPinYinName = pinyin.get(modelName, format='strip')
if os.path.exists("model/" + modelPinYinName):
shutil.rmtree("model/" + modelPinYinName)
if os.path.exists("data/" + modelPinYinName):
shutil.rmtree("data/" + modelPinYinName)
return {"code": 1, "message": "ok", "data": "ok"}
def get(self):
params = self.parser.parse_args()
tagDatas = params.get('tagDatas')
modelName = params.get('modelName')
modelPinYinName = pinyin.get(modelName, format='strip')
with open("data/" + modelPinYinName + "/train.tsv",
'a+',
encoding="utf8") as f:
for text in json.loads(tagDatas):
f.write(text + '\n')
return {"code": 1, "message": "ok", "data": "ok"}
def get(self):
params = self.parser.parse_args()
file_name = pinyin.get(params.get('modelName'), format='strip')
data = file_page_query.page_list(file_name, int(params.get('page')),
int(params.get('count')))
return {
"code": 1,
"message": "ok",
"data": {
"data": data,
"size": file_page_query.count_file_lines(file_name) - 1
}
}
def get(self):
params = self.parser.parse_args()
modelName = params.get('modelName')
data = params.get('data')
# data = [["有保障"],
# ["无风险"],
# ["基金过往数据并不代表未来趋势"],
# ["为什么"],
# ["周杰伦"],
# ]
modelPinYinName = pinyin.get(modelName, format='strip')
if not os.path.exists("model/" + modelPinYinName):
return {"code": 0, "message": "model not exist", "data": ""}
request_data = json.loads(data)
predict_result = predict.predict_tag(modelPinYinName, request_data)
api_result = {}
for i in range(len(request_data)):
api_result[request_data[i][0]] = predict_result[i]
return {"code": 1, "message": "ok", "data": api_result}
def jielong(cityname,list):
returncitylist = []
pinyinlist = []
for i in list:
citypinyin = pinyin.get(i, format='strip', delimiter=' ').split()
pinyinlist.append(citypinyin)
pinyinname = pinyinlist[list.index(cityname)]
lastcityname =pinyinname[-1]
run = True
while run:
usedlist = []
for j in pinyinlist:
if j not in usedlist and j[0] == lastcityname and j != pinyinname:
usedlist.append(j)
returncitylist.append(list[pinyinlist.index(j)])
lastcityname = j[-1]
else:
run = False
return returncitylist
def review_extraction(self):
self._sync_button(doing=self.review_extraction_button)
try:
segments = self._info_handler.get(INFO_TAGS)
words = frozenset(tag for articles in segments.values()
for article in articles.values()
for tag in article)
SelectionDialog(master=self.master,
title="查看 抽取词汇 ({}个)".format(len(words)),
default_value=sorted(
list(words),
key=lambda w: pinyin.get(
w, format="strip", delimiter=" ")),
editable=False)
except ValueError as v:
s = str(v)
self.tracker.log("值错误！ [error='{}']".format(s),
tp=TRACKER_LOG_ERROR,
exc_info=v,
prt=True)
messagebox.showerror(parent=self.master,
title="值错误",
message="值错误！\n"
"请检查对应域是否填写错误！\n"
"\n"
"错误信息:\n{}".format(s))
except Exception as e:
s = str(e)
self.tracker.log("未知错误！ [error='{}']".format(s),
tp=TRACKER_LOG_ERROR,
exc_info=e,
prt=True)
messagebox.showerror(parent=self.master,
title="未知错误",
message="未知错误！\n"
"请联系开发者并汇报该错误！\n"
"\n"
"错误信息:\n{}".format(s))
finally:
self._sync_button()
def write_excel(path, total_summary: Dict[str, Dict[str, Any]],
detail_summary: Dict[str, Dict[str, Dict[str, int]]], sorting,
tracker, show_detail):
"""
Total summary:
{ tag: { detail } }
Detail summary:
{ category: { article: { tag: int } }
"""
tracker.log("正在创建 Excel 文件", prt=True)
# first write the summary page
workbook = xlsxwriter.Workbook(path)
# some format
title_format = workbook.add_format({"bold": True})
# create the summary worksheet
summary = workbook.add_worksheet(name="总结")
# init accum
tracker.init_ticker(
" 进程", "正在输出", 0,
len(total_summary) +
sum(len(tag_details) for tag_details in detail_summary.values()))
row, col = 0, 0
max_width = []
summary.write_string(row, col, "词汇", cell_format=title_format)
max_width.append(2 * WIDTH_RATIO)
col += 1
summary.write_string(row, col, "趋势标签", cell_format=title_format)
max_width.append(4 * WIDTH_RATIO)
col += 1
if show_detail:
summary.write_string(row, col, "拟合系数", cell_format=title_format)
max_width.append(4 * WIDTH_RATIO)
col += 1
summary.write_string(row, col, "RMSE", cell_format=title_format)
max_width.append(4 * WIDTH_RATIO)
col += 1
# write header
cat_header = sorted(sorting.items(), key=lambda i: i[1])
for category, _ in cat_header:
summary.write_string(row, col, category, cell_format=title_format)
max_width.append(len(category) * WIDTH_RATIO)
col += 1
row += 1
col = 0
# write values
for tag, detail in total_summary.items():
tracker.update_disc_fill("写入总结 {}".format(tag))
tracker.tick()
# write tag
summary.write_string(row, col, tag)
max_width[col] = max(len(tag) * WIDTH_RATIO, max_width[col])
col += 1
# write label
label = detail["label"]
if label is not None:
label_name = TREND_NAME[label]
summary.write_string(row, col, label_name)
max_width[col] = max(len(label_name) * WIDTH_RATIO, max_width[col])
else:
summary.write_string(row, col, "-")
max_width[col] = max(1, max_width[col])
col += 1
# write detail
if show_detail:
lr = detail["regression"]
if lr is not None:
coefficient, _, RMSE = lr
summary.write_number(row, col, coefficient)
max_width[col] = max(len(str(coefficient)), max_width[col])
col += 1
summary.write_number(row, col, RMSE)
max_width[col] = max(len(str(RMSE)), max_width[col])
col += 1
else:
summary.write_string(row, col, "-")
max_width[col] = max(1, max_width[col])
col += 1
summary.write_string(row, col, "-")
max_width[col] = max(1, max_width[col])
col += 1
# write
occurrence = detail["occurrence"]
for occ in occurrence:
summary.write_number(row, col, occ)
max_width[col] = max(len(str(occ)), max_width[col])
col += 1
row += 1
col = 0
for index, width in enumerate(max_width):
summary.set_column(index, index, width=width)
# write detail
# { category: { tag: { article: int } }
for category, _ in cat_header:
tag_details = detail_summary[category]
work_sheet = workbook.add_worksheet(category)
row, col = 0, 0
max_width = []
work_sheet.write_string(row, col, "词汇", cell_format=title_format)
max_width.append(2 * WIDTH_RATIO)
col += 1
# write header
header = sorted(
next(iter(tag_details.values())).keys(),
key=lambda n: pinyin.get(n, format="strip", delimiter=" "))
# write header
for name in header:
work_sheet.write_string(row, col, name, cell_format=title_format)
max_width.append(len(name) * WIDTH_RATIO)
col += 1
row += 1
col = 0
# write values
for tag, art_detail in tag_details.items():
tracker.update_disc_fill("写入细节 {}".format(tag))
tracker.tick()
work_sheet.write_string(row, col, tag)
max_width[col] = max(len(tag) * WIDTH_RATIO, max_width[col])
col += 1
for name in header:
val = art_detail[name]
work_sheet.write_number(row, col, val)
max_width[col] = max(len(str(val)), max_width[col])
col += 1
row += 1
col = 0
for index, width in enumerate(max_width):
work_sheet.set_column(index, index, width=width)
tracker.log("正在关闭 Excel 文件", prt=True)
workbook.close()
list6.extend(list5)
for i in list6:
list2 = [k for k in i]
if '-' in list2:
num = list2.index('-')
del list2[:num + 1]
if '：' in list2:
num = list2.index('：')
del list2[:num + 1]
for j in key:
if j in list2:
num = list2.index(j)
del list2[num:]
name = ''
for l in list2:
name = name + l
list3.append(name)
list4 = [i for i in list3 if len(i) >= 2]
citylist = list4[10:]
pinyinlist = []
for i in citylist:
citypinyin = pinyin.get(i, format='strip', delimiter=' ').split()
pinyinlist.append(citypinyin)
cityname = input('请输入城市名：')
lastcityname = pinyinlist[citylist.index(cityname)][-1]
for i in pinyinlist:
if i[0] == lastcityname:
print(citylist[pinyinlist.index(i)])
lastcityname = i[-1]
def myPinyin(strings):
'''
将汉字变为pinyin
'''
return pinyin.get(strings, format="strip",
delimiter="").replace('（', '_').replace('）', '_')
prov_list = rc.findall(html)
# 获取所有的市
rc = re.compile('\<dd.*?\>(.*?)\<\/dd\>')
city_list = rc.findall(html)
# data
data = []
data.append({"en": "beijing", "cn": "北京"})
data.append({"en": "tianjin", "cn": "天津"})
data.append({"en": "shanghai", "cn": "上海"})
data.append({"en": "chongqing", "cn": "重庆"})
for k in range(len(prov_list)):
prov_dict = {}
prov_dict['en'] = pinyin.get(prov_list[k])
prov_dict['cn'] = prov_list[k]
if k == 0 or k == len(prov_list) - 1:
continue
# 获取该省下的所有市
rc = re.compile('\<a.*?\>(.*?)\<\/a\>')
item_city = rc.findall(city_list[k])
prov_dict['city'] = item_city
data.append(prov_dict)
for v in data:
print('*' * 10)
print("该省%s（%s）" % (v['cn'], v['en']))
def setChinese(self, chinese):
self.chinese = chinese
self.pinyin = pinyin.get(chinese)
def __init__(self, english, chinese):
self.english = english
self.chinese = chinese
self.pinyin = pinyin.get(chinese)
#!/bin/python2.7
#coding=utf-8
import os, sys
sys.path.append(os.path.dirname(os.getcwd()))
from models.empire.linkageModel import LinkageModel
from pinyin import pinyin
obj = LinkageModel("/data/cap/python/python_scripts/config/base.conf")
query = obj.getQuery()
num = 0
for item in query:
if item.parent_id != 18:
continue
pinyinStr = pinyin.get(item.linkage_name, "", format="strip")
item.pinyin = pinyinStr
print item.linkage_name
obj.session.commit()
continue
result.append({
"id": index,
"text": line_str[0],
"label": line_str[1].replace("\n", ""),
"exist": 1
})
# for i in range(len(result)):
# print(result[i])
return result
def modify_line(file_path, line, content):
file_path = "data/" + file_path + "/train.tsv"
line += 1
with open(file_path, 'r+', encoding='utf8') as f:
cont = 1
while True:
cont += 1
f.readline()
if cont == int(line):
f.seek(f.tell())
f.write(content + "\n")
break
if __name__ == '__main__':
# page_list('data/fund_risk/train.tsv', 1)
# modify_line('data/fund_risk/train.tsv', 1, "高收益~1~1")
print(pinyin.get("诋毁其他基金管理人", format='strip'))
def get(self):
params = self.parser.parse_args()
file_page_query.modify_line(
pinyin.get(params.get('modelName'), format='strip'),
int(params.get('id')), params.get('data'))
return {"code": 1, "message": "ok", "data": "ok"}
def get(self):
params = self.parser.parse_args()
modelName = params.get('modelName')
modelPinYinName = pinyin.get(modelName, format='strip')
shutil.copytree('data/base_data', 'data/' + modelPinYinName)
return {"code": 1, "message": "ok", "data": modelPinYinName}
from pinyin import pinyin
# 汉字中的间隔符delimiter
a = pinyin.get('杨强', format='strip', delimiter="")
print(a)
# 中间有空格
b = pinyin.get_initial('杨w强')
print(b.replace(' ', ''))
