<?xml version="1.0" encoding="utf-8" ?><rss version="2.0"><channel><title><![CDATA[宁静致远]]></title><description><![CDATA[]]></description><link>https://blog.youkuaiyun.com/weiyudang11</link><language>zh-cn</language><generator>https://blog.youkuaiyun.com/</generator><copyright><![CDATA[Copyright &copy; weiyudang11]]></copyright><item><title><![CDATA[浅谈GCN]]></title><link>https://blog.youkuaiyun.com/weiyudang11/article/details/92662420</link><guid>https://blog.youkuaiyun.com/weiyudang11/article/details/92662420</guid><author>weiyudang11</author><pubDate>Mon, 17 Jun 2019 17:43:37 +0800</pubDate><description><![CDATA[GCN(Graph convolution Network)是Convets在图结构上的自然推广。卷积神经网络是采用局部感知区域、共享权值和空间域上的降采样，相对于位移、缩放和扭曲，具有稳定不变的特性，能够很好的提取图像的空间特征。图结构不具备图片的平移不变性，传统的卷积方式不适用于图结构。图中每个节点的邻域节点数目不一致，无法用同样尺寸的卷积核进行提取特。



而GCN的本质就是提取图的结构...]]></description><category></category></item><item><title><![CDATA[py2neo3.0 批处理方案]]></title><link>https://blog.youkuaiyun.com/weiyudang11/article/details/78808429</link><guid>https://blog.youkuaiyun.com/weiyudang11/article/details/78808429</guid><author>weiyudang11</author><pubDate>Thu, 14 Dec 2017 23:10:22 +0800</pubDate><description><![CDATA[py2neo 批处理方案neo4j是是一个高性能的NOSQL图形数据库，它将结构化的数据存储在网络上而不是表中。它是一个嵌入式的，基于磁盘的、具备完备的事务特性的JAVA持续化引擎，但是它将结构化数据存储在网络(从数学角度叫做图)上而不是表中。neo4j 建立数据库有多种方式：
cypher 
load csv 
REST API
第一种方式不适合批量导入，第二种需要前期的处理，因为比较喜欢pyth]]></description><category></category></item><item><title><![CDATA[python 中的hash]]></title><link>https://blog.youkuaiyun.com/weiyudang11/article/details/52684495</link><guid>https://blog.youkuaiyun.com/weiyudang11/article/details/52684495</guid><author>weiyudang11</author><pubDate>Tue, 27 Sep 2016 20:43:05 +0800</pubDate><description><![CDATA[simhashimport re 
from simhash import Simhashdef get_features(s):
    withd=4
    s=s.lower()
    s=re.sub(r'[^\w]+','',s)
    return [s[i:i+withd] for i in range(max(len(s)-withd+1,1))]Simhash(get_fea]]></description><category></category></item><item><title><![CDATA[空间数据聚类  DBSCAN]]></title><link>https://blog.youkuaiyun.com/weiyudang11/article/details/52684333</link><guid>https://blog.youkuaiyun.com/weiyudang11/article/details/52684333</guid><author>weiyudang11</author><pubDate>Tue, 27 Sep 2016 20:22:05 +0800</pubDate><description><![CDATA[# 地理空间数据聚类
%matplotlib  inline
import numpy as np,pandas as pd,matplotlib.pyplot as plt
from sklearn.cluster import DBSCAN
from  geopy.distance import great_circle
from shapely.geometry  import MultiPo]]></description><category></category></item><item><title><![CDATA[Python 文件操作]]></title><link>https://blog.youkuaiyun.com/weiyudang11/article/details/52684111</link><guid>https://blog.youkuaiyun.com/weiyudang11/article/details/52684111</guid><author>weiyudang11</author><pubDate>Tue, 27 Sep 2016 19:49:28 +0800</pubDate><description><![CDATA[File模块  文档filepath=r"E:\workpalce\Python\data\Andersen's_Fairy_Tales\100.txt"
f=file(filepath,mode='a+')### 模式f.mode### 判断是否关闭f.closed### 编码f.encoding### 终端是否能访问f.isatty()f.fileno()f.namef.readline()##]]></description><category></category></item><item><title><![CDATA[geopandas 的使用以及相关问题]]></title><link>https://blog.youkuaiyun.com/weiyudang11/article/details/52683877</link><guid>https://blog.youkuaiyun.com/weiyudang11/article/details/52683877</guid><author>weiyudang11</author><pubDate>Tue, 27 Sep 2016 19:16:23 +0800</pubDate><description><![CDATA[geopandas 的使用以及相关问题Geoff BoeingMichelle Fullwood%matplotlib  inline
import pandas as pd
import geopandas as gpd
from geopandas import GeoDataFrame, read_file
from geopandas.tools import sjoin
from shap]]></description><category></category></item><item><title><![CDATA[Python  空间数据处理]]></title><link>https://blog.youkuaiyun.com/weiyudang11/article/details/52495645</link><guid>https://blog.youkuaiyun.com/weiyudang11/article/details/52495645</guid><author>weiyudang11</author><pubDate>Sat, 10 Sep 2016 16:32:52 +0800</pubDate><description><![CDATA[Geopy测试GeoCodeing:得出具体的地址from geopy.geocoders import Nominatim
geolocator = Nominatim()
location = geolocator.geocode("中国人民大学")
print(location.address)中国人民大学, 人民大学北路, 稻香园南社区, 海淀区, 北京市, 100872, 中国
经纬度信息]]></description><category></category></item><item><title><![CDATA[神经网络]]></title><link>https://blog.youkuaiyun.com/weiyudang11/article/details/52469344</link><guid>https://blog.youkuaiyun.com/weiyudang11/article/details/52469344</guid><author>weiyudang11</author><pubDate>Thu, 08 Sep 2016 11:18:02 +0800</pubDate><description><![CDATA[交叉熵代价函数（作用及公式推导）： 
http://blog.youkuaiyun.com/u014313009/article/details/51043064 
WILDML：http://www.wildml.com/2015/09/implementing-a-neural-network-from-scratch/]]></description><category></category></item><item><title><![CDATA[各种熵]]></title><link>https://blog.youkuaiyun.com/weiyudang11/article/details/52468587</link><guid>https://blog.youkuaiyun.com/weiyudang11/article/details/52468587</guid><author>weiyudang11</author><pubDate>Thu, 08 Sep 2016 10:29:18 +0800</pubDate><description><![CDATA[自信息量I(x)=-log(p(x))，其他依次类推。离散变量x的熵H(x)=E(I(x))=-∑xp(x)lnp(x)\sum\limits_{x}{p(x)lnp(x)}连续变量x的微分熵H(x)=E(I(x))=-∫p(x)lnp(x)dx\int{p(x)lnp(x)dx} 条件熵H(y|x)=-∫∫p(x,y)lnp(y|x)dydx\int\int{p(x,y)lnp(y|x)dydx}]]></description><category></category></item><item><title><![CDATA[短文本相似度度量]]></title><link>https://blog.youkuaiyun.com/weiyudang11/article/details/52402867</link><guid>https://blog.youkuaiyun.com/weiyudang11/article/details/52402867</guid><author>weiyudang11</author><pubDate>Thu, 01 Sep 2016 19:46:54 +0800</pubDate><description><![CDATA[短文本语义相似度度量基于深度学习的短文的语义相似度计算依存句法分析依存句法分析是一种常用的句子结构的表达方式。在依存句法中，句子的结构是以其谓语为中心，词语之间的相互依存而构成的树形结构。依存句法分析可以获得词语间的修饰关系，有助于分析句子的结构及其所报含的信息。 
常用的依存句法分析模型可以分为两个大类
基于图模型，比较常见的实现是哈工大的LTP以及MTPParser
基于转移模型，具有线性时间复]]></description><category></category></item><item><title><![CDATA[Scipy-矩阵运算]]></title><link>https://blog.youkuaiyun.com/weiyudang11/article/details/52369488</link><guid>https://blog.youkuaiyun.com/weiyudang11/article/details/52369488</guid><author>weiyudang11</author><pubDate>Tue, 30 Aug 2016 15:10:54 +0800</pubDate><description><![CDATA[Scipy-矩阵运算：http://blog.sina.com.cn/s/blog_70586e000100moen.html]]></description><category></category></item><item><title><![CDATA[最优化导论1]]></title><link>https://blog.youkuaiyun.com/weiyudang11/article/details/52355562</link><guid>https://blog.youkuaiyun.com/weiyudang11/article/details/52355562</guid><author>weiyudang11</author><pubDate>Mon, 29 Aug 2016 18:32:38 +0800</pubDate><description><![CDATA[证明方法A⇒BA\Rightarrow B就是组合命题“（非A）或B”，经常读作“A当且仅当Ｂ”，“如果Ａ，则Ｂ”、“Ａ是Ｂ的充分条件”或者“Ｂ是Ａ的必要条件”。可以将两个命题组合成等价命题的形式A⇐⇒B A\Leftarrow \Rightarrow B 读作“A成立，当且仅当B成立”或者“A等价于B”或者“A对于B是充分必要的”向量与矩阵
  
  如果方程α1a1+α2a2+⋅⋅⋅⋅⋅+αka]]></description><category></category></item><item><title><![CDATA[共轭梯度下降]]></title><link>https://blog.youkuaiyun.com/weiyudang11/article/details/52335735</link><guid>https://blog.youkuaiyun.com/weiyudang11/article/details/52335735</guid><author>weiyudang11</author><pubDate>Sat, 27 Aug 2016 16:26:20 +0800</pubDate><description><![CDATA[共轭梯度下降：https://en.wikipedia.org/wiki/Conjugate_gradient_method]]></description><category></category></item><item><title><![CDATA[theano学习指南1]]></title><link>https://blog.youkuaiyun.com/weiyudang11/article/details/52330827</link><guid>https://blog.youkuaiyun.com/weiyudang11/article/details/52330827</guid><author>weiyudang11</author><pubDate>Fri, 26 Aug 2016 18:16:35 +0800</pubDate><description><![CDATA[http://www.cnblogs.com/xueliangliu/archive/2013/04/03/2997437.html]]></description><category></category></item><item><title><![CDATA[python  库知识]]></title><link>https://blog.youkuaiyun.com/weiyudang11/article/details/52330312</link><guid>https://blog.youkuaiyun.com/weiyudang11/article/details/52330312</guid><author>weiyudang11</author><pubDate>Fri, 26 Aug 2016 17:21:22 +0800</pubDate><description><![CDATA[推荐11个Python 库：http://www.jb51.net/article/60255.htm]]></description><category></category></item><item><title><![CDATA[利用代理服务器的简单爬虫程序]]></title><link>https://blog.youkuaiyun.com/weiyudang11/article/details/52327834</link><guid>https://blog.youkuaiyun.com/weiyudang11/article/details/52327834</guid><author>weiyudang11</author><pubDate>Fri, 26 Aug 2016 14:06:39 +0800</pubDate><description><![CDATA[简单的爬虫，用到的知识点： 
1、代理服务器 
2、多线程 
需要提高的： 
1、分布式机制#encoding=utf8
import os
from bs4 import BeautifulSoup
import urllib2
from multiprocessing import Pool
import random
def proxys():
    proxy_cont=open("pro]]></description><category></category></item><item><title><![CDATA[python mysql 插入]]></title><link>https://blog.youkuaiyun.com/weiyudang11/article/details/52313813</link><guid>https://blog.youkuaiyun.com/weiyudang11/article/details/52313813</guid><author>weiyudang11</author><pubDate>Thu, 25 Aug 2016 13:39:05 +0800</pubDate><description><![CDATA[import MySQLdb
conn=MySQLdb.connect(host="localhost",user="root",passwd="19900410w",port=3306,db="crawl")
cur=conn.cursor()
sql="insert into proxy_avi values(%s,%s)"
# sql="show tables"
# sql="select *]]></description><category></category></item><item><title><![CDATA[urllib2 使用代理服务器]]></title><link>https://blog.youkuaiyun.com/weiyudang11/article/details/52313407</link><guid>https://blog.youkuaiyun.com/weiyudang11/article/details/52313407</guid><author>weiyudang11</author><pubDate>Thu, 25 Aug 2016 12:11:43 +0800</pubDate><description><![CDATA[相关博客： 
Python 爬虫入门（二）—— IP代理使用： 
http://www.cnblogs.com/hearzeus/p/5157016.html 
python爬虫如何抓取代理服务器： 
https://segmentfault.com/n/1330000005070016# The proxy address and port:
proxy_info = { 'host' : 'pr]]></description><category></category></item><item><title><![CDATA[文本分析-gensim]]></title><link>https://blog.youkuaiyun.com/weiyudang11/article/details/52300632</link><guid>https://blog.youkuaiyun.com/weiyudang11/article/details/52300632</guid><author>weiyudang11</author><pubDate>Wed, 24 Aug 2016 15:08:48 +0800</pubDate><description><![CDATA[from sklearn.feature_extraction.text  import CountVectorizer,TfidfVectorizer
from gensim import corpora,similarities,models
import jieba
from prettyprint import pp读取诛仙文本内容zhuxian=open('zhuxian.txt').re]]></description><category></category></item><item><title><![CDATA[统计学之回归分析]]></title><link>https://blog.youkuaiyun.com/weiyudang11/article/details/52239559</link><guid>https://blog.youkuaiyun.com/weiyudang11/article/details/52239559</guid><author>weiyudang11</author><pubDate>Thu, 18 Aug 2016 11:50:49 +0800</pubDate><description><![CDATA[统计学-回归分析回归分析]]></description><category></category></item></channel></rss>