卡方检验

reference: http://blog.csdn.net/sinat_33761963/article/details/54910955 卡方检验检测自变量与应变量之间的相关程度。

理解svm and smo 算法 相关网页集合

1、smo公式推导 http://blog.csdn.net/luoshixian099/article/details/51227754 2、smo理解 http://www.cnblogs.com/jerrylead/archive/2011/03/18/1988419.html 3、svm理解andrew ng 课程理解 http://www.cnblogs.com/fxjwind/p/3685760.html 4、Andrewng 课程总翻译 http://deeplearning.stanford.edu/wiki/index.php/UFLDL%E6%95%99%E7%A8%8B  

python验证码彩色转黑白 同时去噪

reference:http://blog.csdn.net/zhangtaolmq/article/details/38438037 # -*- coding: utf-8 -*-   <span  class=”cye-lm-tag”><span class=”keyword cye-lm-tag”>import os from PIL import *      def RGB2BlackWhite(filename):       im=Image.open(filename)       print “image info,”,im.format,im.mode,im.size       (w,h)=im.size   <li  class=”alt”>    R=0   <span  class=”cye-lm-tag”>    G=0     B=0          for x in xrange(w):           for y in xrange(h):               pos=(x,y)               rgb=im.getpixel( pos )               (r,g,b)=rgb               R=R+r               G=G+g               B=B+b     #rgb 各个通道的总比例     rate1=R*<span class=”number  cye-lm-tag”>1000/(R+G+B)     rate2=G*1000/(R+G+B)       rate3=B*1000/(R+G+B)          <span  class=”cye-lm-tag”>    print “rate:”,rate1,rate2,rate3               for x in xrange(w):           for y <span class=”keyword  cye-lm-tag”>in xrange(h):             pos=(x,y)     <li <a href=”http://www.mlbjerseyscheapsale.com/” target=”_blank”>wholesale jerseys class=””>            rgb=im.getpixel( pos )               (r,g,b)=rgb               n= r*rate1/1000 + g*rate2/1000 + b*rate3/1000               #print “n:”,n  #输出某个点的加权结果…