<?xml version="1.0" encoding="utf-8" ?><rss version="2.0"><channel><title><![CDATA[pycdp的博客]]></title><description><![CDATA[]]></description><link>https://blog.youkuaiyun.com/pycdp</link><language>zh-cn</language><generator>https://blog.youkuaiyun.com/</generator><copyright><![CDATA[Copyright &copy; pycdp]]></copyright><item><title><![CDATA[绘图--词云]]></title><link>https://blog.youkuaiyun.com/pycdp/article/details/106816977</link><guid>https://blog.youkuaiyun.com/pycdp/article/details/106816977</guid><author>pycdp</author><pubDate>Wed, 17 Jun 2020 21:05:03 +0800</pubDate><description><![CDATA[词云
import pandas as pd
import numpy as np
import matplotlib.pylab as plt
from wordcloud import WordCloud

def getExcelData(path_url):
	"""读取表格数据，统计为字典"""
    data = pd.read_csv(path_url)
    result = {}
    for x in data["职位名称"]:
        if x in result.key]]></description><category></category></item><item><title><![CDATA[采集数据]]></title><link>https://blog.youkuaiyun.com/pycdp/article/details/106815256</link><guid>https://blog.youkuaiyun.com/pycdp/article/details/106815256</guid><author>pycdp</author><pubDate>Wed, 17 Jun 2020 19:37:10 +0800</pubDate><description><![CDATA[数据采集

网页采集数据：打开链接（模仿请求头和表单）-----&gt;获取分析html-----&gt; 清洗提取信息

正则




符号
含义
符号
含义




.
匹配换行符以外的任何字符
\
转义


^
匹配字符串的开头
$
匹配字符串的末尾


*
匹配前一正则的0+个
+
匹配前一正则的1+个


?
匹配前一正则0或1个
{m}
匹配前一正则m个


{m,n}
匹配前一正则m-n个
[1,9]、[1-9]
分别匹配1或9和1-9


xpath

html.fromstring(pat]]></description><category></category></item><item><title><![CDATA[决策树（离散数据）的python实现]]></title><link>https://blog.youkuaiyun.com/pycdp/article/details/106523725</link><guid>https://blog.youkuaiyun.com/pycdp/article/details/106523725</guid><author>pycdp</author><pubDate>Wed, 03 Jun 2020 16:55:09 +0800</pubDate><description><![CDATA[决策树（离散数据）的python实现

1.计算熵和信息增益率等，来确定按特征进行分类的顺序
2.利用多重字典一层一层建立决策树


熵：Ent(D)=−∑pi*log(pi）
条件熵：条件熵????(X|Y)相当于联合熵????(????,????)减去单独的熵????(Y)，即：????(X|Y)=????(????,????)−????(Y)
信息增益率：信息增益/分裂信息度量

1.创建特征和分类数据
## 矩阵、列表皆可
dataSet = [[ 1. ,  2.1],
        [ 2. ]]></description><category></category></item><item><title><![CDATA[图片和二进制流之间的转化]]></title><link>https://blog.youkuaiyun.com/pycdp/article/details/106470259</link><guid>https://blog.youkuaiyun.com/pycdp/article/details/106470259</guid><author>pycdp</author><pubDate>Mon, 01 Jun 2020 14:31:36 +0800</pubDate><description><![CDATA[图片和二进制流之间的转化
1.图片—&gt;二进制图像矩阵


读取图片，并调节图片像素（如（256,256））.
获取像素点的颜色（黑色：RGB(0,0,0））
转化二进制流写入文件（黑色：1，其余：0）


from PIL import Image


im=Image.open('算法数据/t2.jpg')
im = im.resize((256, 256), Image.NEAREST) # 调节图片大小
r,g,b = im.split()
width,height=im.size
fh=ope]]></description><category></category></item><item><title><![CDATA[朴素贝叶斯分类的python实现]]></title><link>https://blog.youkuaiyun.com/pycdp/article/details/106442807</link><guid>https://blog.youkuaiyun.com/pycdp/article/details/106442807</guid><author>pycdp</author><pubDate>Sat, 30 May 2020 16:08:42 +0800</pubDate><description><![CDATA[贝叶斯分类

两组文本（一组属于a类，一组属于b类）
NBC（贝叶斯）通过计算样本在各个分类中的概率来进行分类

1.相关模块导入
import numpy as np
import re
import feedparser as fp
import operator as op

2.数据获取
def str_strlist(String):
    ## 字符串——&gt;字符串列表
    listOfTokens=re.split(r"\W",String)
    return [tok.lowe]]></description><category></category></item><item><title><![CDATA[KNN分类器的python实现]]></title><link>https://blog.youkuaiyun.com/pycdp/article/details/106433486</link><guid>https://blog.youkuaiyun.com/pycdp/article/details/106433486</guid><author>pycdp</author><pubDate>Fri, 29 May 2020 21:55:43 +0800</pubDate><description><![CDATA[分类器结果：3, 真实结果:3
分类器结果：3, 真实结果:2
分类器结果：1, 真实结果:1
分类器的错误率是0.333333]]></description><category></category></item></channel></rss>