算法调参 weight_ratio, weight_seqratio
from openpyxl import Workbook
import xlrd
import time
import Levenshtein as Le target_city_list = ['北京市', '上海市', '深圳市', '广州市']
source_name = 'JMTool任务_csv_py_wholeCSV-加百度170826165729'
BDpoi_list_tag, BDpoi_list_tagb = '|-|', '|--|'
FEXCEL = '%s%s' % (source_name, '.xlsx') weight_ratio, weight_seqratio = 0.7, 0.3 def main_():
global source_name
data = xlrd.open_workbook(FEXCEL)
table = data.sheets()[0]
nrows, ncols = table.nrows, table.ncols
res_dic = {}
for i in range(0, nrows):
l = table.row_values(i)
dbid, area_code, ref_area_type_code, city, district, address, city_street, name_, BDpoi_list = l
if dbid == 'dbid':
continue
if city not in target_city_list:
continue
if city not in res_dic:
res_dic[city] = {}
if district not in res_dic[city]:
res_dic[city][district] = {}
if name_ not in res_dic[city][district]:
res_dic[city][district][name_] = [] if BDpoi_list.find(BDpoi_list_tag) == -1:
ll = dbid, area_code, ref_area_type_code, city, district, address, city_street, name_, BDpoi_list, '', '', ''
res_dic[city][district][name_].append(ll)
else:
addr_ = '%s%s%s%s' % (city, district, address, city_street)
chk_name_lsit, cmp_list, sorted_ratio_seqratio_res_dic = [name_, addr_], BDpoi_list.split(
BDpoi_list_tag), {}
for ii in cmp_list:
if len(ii) == 0:
continue
cmp_, BD_name, BD_addr = ['', ''], '', ''
cmp_one = ii.split(BDpoi_list_tagb)
if len(cmp_one) == 2:
# format data -fair
BD_name, BD_addr = cmp_[0], cmp_[1] = cmp_one[0], cmp_one[1].replace(city, '').replace(district, '')
else:
BD_name = cmp_[0] = cmp_one[0]
ratio_res, seqratio_res = Le.ratio(name_, BD_name), Le.seqratio(chk_name_lsit, cmp_)
ratio_seqratio_res = weight_ratio * ratio_res + weight_seqratio * seqratio_res
ll = dbid, area_code, ref_area_type_code, city, district, address, city_street, name_, BDpoi_list, BD_name, BD_addr, ratio_seqratio_res, ratio_res, seqratio_res
if ratio_seqratio_res not in sorted_ratio_seqratio_res_dic:
sorted_ratio_seqratio_res_dic[ratio_seqratio_res] = []
sorted_ratio_seqratio_res_dic[ratio_seqratio_res].append(ll)
sorted_seqratio_res_list = sorted(sorted_ratio_seqratio_res_dic)
for ratio_seqratio_res in sorted_seqratio_res_list:
lll = sorted_ratio_seqratio_res_dic[ratio_seqratio_res]
for vl in lll:
res_dic[city][district][name_].append(vl) wb = Workbook()
worksheet = wb.active
file_title_str = ' dbid, area_code, ref_area_type_code, city, district, address, city_street, name_, BDpoi_list, BD_name, BD_addr, ratio_seqratio_res, ratio_res, seqratio_res'
file_title_l = file_title_str.replace(' ', '').split(',')
worksheet.append(file_title_l)
for city in res_dic:
for district in res_dic[city]:
for name_ in res_dic[city][district]:
l = res_dic[city][district][name_]
for ll in l:
worksheet.append(ll)
localtime_ = time.strftime("%y%m%d%H%M%S", time.localtime())
file_name = '%s%s%s' % (source_name, '-Levenshtein', localtime_)
file_name_save = '%s%s' % (file_name, '.xlsx')
wb.save(file_name_save) wb = Workbook()
worksheet = wb.active
file_title_str = ' dbid, area_code, ref_area_type_code, city, district, address, city_street, name_,BDpoi_list,max_BD_name, max_BD_addr, max_ratio_seqratio_res, ratio_res, seqratio_res'
file_title_l = file_title_str.replace(' ', '').split(',')
worksheet.append(file_title_l)
for city in res_dic:
for district in res_dic[city]:
for name_ in res_dic[city][district]:
l = res_dic[city][district][name_]
lll = l[-1] worksheet.append(lll)
localtime_ = time.strftime("%y%m%d%H%M%S", time.localtime())
file_name = '%s%s%s' % (file_name, '-Levenshtein-ordered', localtime_)
file_name_save = '%s%s' % (file_name, '.xlsx')
wb.save(file_name_save) main_()
from openpyxl import Workbook
import xlrd
import time
import Levenshtein as Le target_city_list = ['深圳市']
BDpoi_list_tag, BDpoi_list_tagb = '|-|', '|--|' source_name = 'JMTool任务_csv_py_wholeCSV_住宅小区-加百度170826152533'
FEXCEL = '%s%s' % (source_name, '.xlsx')
weight_ratio, weight_seqratio = 0.7, 0.3 def main_():
global source_name
data = xlrd.open_workbook(FEXCEL)
table = data.sheets()[0]
nrows, ncols = table.nrows, table.ncols
res_dic = {}
for i in range(0, nrows):
l = table.row_values(i)
dbid, area_code, ref_area_type_code, city, district, address, city_street, name_, name_reduction, BDpoi_list = l
if dbid == 'dbid':
continue
if city not in target_city_list:
continue
if city not in res_dic:
res_dic[city] = {}
if district not in res_dic[city]:
res_dic[city][district] = {}
if name_ not in res_dic[city][district]:
res_dic[city][district][name_] = [] if BDpoi_list.find(BDpoi_list_tag) == -1:
ll = dbid, area_code, ref_area_type_code, city, district, address, city_street, name_, name_reduction, BDpoi_list, '', '', ''
res_dic[city][district][name_].append(ll)
else:
addr_ = '%s%s%s%s' % (city, district, address, city_street)
chk_name_lsit, cmp_list, sorted_ratio_seqratio_res_dic = [name_reduction, addr_], BDpoi_list.split(
BDpoi_list_tag), {}
for ii in cmp_list:
if len(ii) == 0:
continue
cmp_, BD_name, BD_addr = ['', ''], '', ''
cmp_one = ii.split(BDpoi_list_tagb)
if len(cmp_one) == 2:
# format data -fair
BD_name, BD_addr = cmp_[0], cmp_[1] = cmp_one[0], cmp_one[1].replace(city, '').replace(district, '')
else:
BD_name = cmp_[0] = cmp_one[0]
ratio_res, seqratio_res = Le.ratio(name_reduction, BD_name), Le.seqratio(chk_name_lsit, cmp_)
ratio_seqratio_res = weight_ratio * ratio_res + weight_seqratio * seqratio_res
ll = dbid, area_code, ref_area_type_code, city, district, address, city_street, name_, name_reduction, BDpoi_list, BD_name, BD_addr, ratio_seqratio_res, ratio_res, seqratio_res
if ratio_seqratio_res not in sorted_ratio_seqratio_res_dic:
sorted_ratio_seqratio_res_dic[ratio_seqratio_res] = []
sorted_ratio_seqratio_res_dic[ratio_seqratio_res].append(ll)
sorted_seqratio_res_list = sorted(sorted_ratio_seqratio_res_dic)
for ratio_seqratio_res in sorted_seqratio_res_list:
lll = sorted_ratio_seqratio_res_dic[ratio_seqratio_res]
for vl in lll:
res_dic[city][district][name_].append(vl) wb = Workbook()
worksheet = wb.active
file_title_str = 'dbid, area_code, ref_area_type_code, city, district, address, city_street, name_, name_reduction, BDpoi_list, BD_name, BD_addr, ratio_seqratio_res, ratio_res, seqratio_res'
file_title_l = file_title_str.replace(' ', '').split(',')
worksheet.append(file_title_l)
for city in res_dic:
for district in res_dic[city]:
for name_ in res_dic[city][district]:
l = res_dic[city][district][name_]
for ll in l:
worksheet.append(ll)
localtime_ = time.strftime("%y%m%d%H%M%S", time.localtime())
file_name = '%s%s%s' % (source_name, '-Levenshtein', localtime_)
file_name_save = '%s%s' % (file_name, '.xlsx')
wb.save(file_name_save) wb = Workbook()
worksheet = wb.active
file_title_str = 'dbid, area_code, ref_area_type_code, city, district, address, city_street, name_, name_reduction,BDpoi_list,max_BD_name, max_BD_addr, max_ratio_seqratio_res, ratio_res, seqratio_res'
file_title_l = file_title_str.replace(' ', '').split(',')
worksheet.append(file_title_l)
for city in res_dic:
for district in res_dic[city]:
for name_ in res_dic[city][district]:
l = res_dic[city][district][name_]
lll = l[-1] worksheet.append(lll)
localtime_ = time.strftime("%y%m%d%H%M%S", time.localtime())
file_name = '%s%s%s' % (file_name, '-Levenshtein-ordered', localtime_)
file_name_save = '%s%s' % (file_name, '.xlsx')
wb.save(file_name_save) main_()
算法调参 weight_ratio, weight_seqratio的更多相关文章
- annoy ANN算法 调参
search_k serach_k越大,越准确,但是要在时间和准确率之间取个trade off During the query it will inspect up to search_k node ...
- CatBoost算法和调参
欢迎关注博主主页,学习python视频资源 sklearn实战-乳腺癌细胞数据挖掘(博主亲自录制视频) https://study.163.com/course/introduction.htm?co ...
- k-近邻算法采用for循环调参方法
//2019.08.02下午#机器学习算法中的超参数与模型参数1.超参数:是指机器学习算法运行之前需要指定的参数,是指对于不同机器学习算法属性的决定参数.通常来说,人们所说的调参就是指调节超参数.2. ...
- 调参、最优化、ml算法（未完成）
最优化方法 调参方法 ml算法 梯度下降gd grid search lr 梯度上升 随机梯度下降 pca 随机梯度下降sgd 贝叶斯调参 lda 牛顿算法 knn 拟牛顿算法 kmeans ...
- scikit-learn随机森林调参小结
在Bagging与随机森林算法原理小结中,我们对随机森林(Random Forest, 以下简称RF)的原理做了总结.本文就从实践的角度对RF做一个总结.重点讲述scikit-learn中RF的调参注 ...
- scikit-learn 梯度提升树(GBDT)调参小结
在梯度提升树(GBDT)原理小结中,我们对GBDT的原理做了总结,本文我们就从scikit-learn里GBDT的类库使用方法作一个总结,主要会关注调参中的一些要点. 1. scikit-learn ...
- word2vec参数调整 及lda调参
一.word2vec调参 ./word2vec -train resultbig.txt -output vectors.bin -cbow 0 -size 200 -window 5 -neg ...
- 漫谈PID——实现与调参
闲话: 作为一个控制专业的学生,说起PID,真是让我又爱又恨.甚至有时候会觉得我可能这辈子都学不会pid了,但是经过一段时间的反复琢磨,pid也不是很复杂.所以在看懂pid的基础上,写下这篇文章,方便 ...
- hyperopt自动调参
hyperopt自动调参 在传统机器学习和深度学习领域经常需要调参,调参有些是通过通过对数据和算法的理解进行的,这当然是上上策,但还有相当一部分属于"黑盒" hyperopt可以帮 ...
随机推荐
- [转]js模块化编程之彻底弄懂CommonJS和AMD/CMD！
原文: https://www.cnblogs.com/chenguangliang/p/5856701.html ------------------------------------------ ...
- Gentoo:startx出现Failed to load module问题
安装完xorg-server后,startx启动桌面环境,出现缺少模块错误. 查看log: cat /var/log/Xorg.0.log | grep EE [75.403] (EE) Failed ...
- Nodejs扩展，实现消息弹窗
參考https://github.com/olalonde/node-notify的实现 模块的C++代码 node_gtknotify.cc #include <v8.h> #inclu ...
- Java 调用存储过程、函数
一.Java调用存储Oracle存储过程 测试用表: --创建用户表 create table USERINFO ( username ) not null, password ) not null ...
- shell脚本通过ping命令来获取平均延时
#!/bin/bash #设置环境变量 PATH="/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/root/bin" exp ...
- rational rose画UML图
原文见:http://blog.csdn.net/cjr15233661143/article/details/8532997 UML是一种建模语言,是系统建模的标准.我们之所以建模是因为大规模的系统 ...
- Android 网络状态的监控
1 http://www.cnblogs.com/qingblog/archive/2012/07/19/2598983.html 2
- Qemu线程池介绍
有时我们希望把一部分工作通过创建线程的方式异步执行,这样我们可以在执行任务的同时,继续执行其他任务.但是如果这种需求比较多的话,频繁的创建和销毁线程带来很大的性能损耗.如果我们能创建一个或一些线程,然 ...
- 官方教程Stealth学习笔记（一）
今天開始要更新官方教程stealth的学习笔记啦, 我将会记录和解说一个小游戏基本的流程和关键地方的技巧. 我会依照官方教程的顺序来更新. ...
- 李洪强经典面试题40-可能碰到的iOS笔试面试题-C语言
可能碰到的iOS笔试面试题(4)--C语言 可能碰到的iOS笔试面试题(4)--C语言 C语言,开发的基础功底,iOS很多高级应用都要和C语言打交道,所以,C语言在iOS开发中的重要性,你懂的.里面的 ...