<?xml version="1.0" encoding="utf-8" ?><rss version="2.0"><channel><title><![CDATA[ginger188的专栏]]></title><description><![CDATA[]]></description><link>https://blog.youkuaiyun.com/winter_evening</link><language>zh-cn</language><generator>https://blog.youkuaiyun.com/</generator><copyright><![CDATA[Copyright &copy; winter_evening]]></copyright><item><title><![CDATA[利用gitlab pages和hexo搭建一个个人博客]]></title><link>https://blog.youkuaiyun.com/winter_evening/article/details/72353953</link><guid>https://blog.youkuaiyun.com/winter_evening/article/details/72353953</guid><author>winter_evening</author><pubDate>Tue, 16 May 2017 22:46:48 +0800</pubDate><description><![CDATA[利用gitlab搭建个人博客]]></description><category></category></item><item><title><![CDATA[BN层计算的为什么不是协方差矩阵的理解]]></title><link>https://blog.youkuaiyun.com/winter_evening/article/details/72083790</link><guid>https://blog.youkuaiyun.com/winter_evening/article/details/72083790</guid><author>winter_evening</author><pubDate>Sun, 14 May 2017 23:12:14 +0800</pubDate><description><![CDATA[为什么协方差矩阵的秩不大于n-1？
为什么BN层计算的时候不用协方差？]]></description><category></category></item><item><title><![CDATA[关于numpy中cov(x)与cover(x,y)的一些理解]]></title><link>https://blog.youkuaiyun.com/winter_evening/article/details/71773502</link><guid>https://blog.youkuaiyun.com/winter_evening/article/details/71773502</guid><author>winter_evening</author><pubDate>Fri, 12 May 2017 21:07:48 +0800</pubDate><description><![CDATA[np.cov(x)与np.cov(x,y)的理解]]></description><category></category></item><item><title><![CDATA[统计学习方法(4)——贝叶斯分类算法]]></title><link>https://blog.youkuaiyun.com/winter_evening/article/details/70232409</link><guid>https://blog.youkuaiyun.com/winter_evening/article/details/70232409</guid><author>winter_evening</author><pubDate>Tue, 18 Apr 2017 20:38:52 +0800</pubDate><description><![CDATA[朴素贝叶斯算法实现简单，学习和预测的效率均很高，是一种非常常用的方法。1 贝叶斯算法贝叶斯算法指通过学习数据的先验概率P(Y)P(Y)和类条件概率P(X=x|Y=ck)P(X=x|Y=c_k)分布，通过贝叶斯定理计算出后验概率P(Y=ck|X=x)P(Y=c_k|X=x)。因为在实际中，我们往往比较容易得到前两者，通过(1)式我们便可以得到我们实际希望得到的样本在满足x的条件下，属于ckc_k这个类]]></description><category></category></item><item><title><![CDATA[统计学习方法(3)——KNN,KD树及其Python实现]]></title><link>https://blog.youkuaiyun.com/winter_evening/article/details/70196080</link><guid>https://blog.youkuaiyun.com/winter_evening/article/details/70196080</guid><author>winter_evening</author><pubDate>Sun, 16 Apr 2017 16:12:52 +0800</pubDate><description><![CDATA[1 k近邻算法k近邻算法是一种基本的分类算法，它的思想非常的简单直观，即一个样本的类别应该和训练数据集中和它距离最近的k个样本中多数样本所属的类别相同，因此，k近邻法分类时没有显式的学习过程。k近邻法的模型实际是一种对特征空间的划分，模型由距离度量、k值的选择和决策规则决定，对于决策规则，我们一般使用多数表决的原则，因此模型的表现主要由距离度量和k值决定。1.1 距离度量特征空间中两点间的距离可以看]]></description><category></category></item><item><title><![CDATA[统计学习方法(2)——感知机原始形式、对偶形式及Python实现]]></title><link>https://blog.youkuaiyun.com/winter_evening/article/details/70196040</link><guid>https://blog.youkuaiyun.com/winter_evening/article/details/70196040</guid><author>winter_evening</author><pubDate>Sun, 16 Apr 2017 16:07:23 +0800</pubDate><description><![CDATA[感知机作为一种最简单的线性二分类模型，可以在输入空间（特征空间）将实例划分为正负两类。本文主要介绍感知机两种形式对应的学习算法及Python实现。　感知机学习算法的原始形式对于输入空间，感知机通过以下函数将其映射至{+1，-1}的输出空间 
f(x)=sign(w⋅x+b)f(x) = sign(w \cdot x + b)(1) 
对于所有的错分类点i∈Mi \in M，都有−yi(w⋅xi+b)]]></description><category></category></item><item><title><![CDATA[统计学习方法(1)——正则化、误差上界的证明等]]></title><link>https://blog.youkuaiyun.com/winter_evening/article/details/70196011</link><guid>https://blog.youkuaiyun.com/winter_evening/article/details/70196011</guid><author>winter_evening</author><pubDate>Sun, 16 Apr 2017 16:04:57 +0800</pubDate><description><![CDATA[First of all,最近打算系统的学习一遍《统计学习方法》这本书，把里面提到的概念和公式都好好学习和推导一遍.因此今后的一系列文章将会记录一些我在阅读本书时的一些心得体会.
统计学习作为数据挖掘的手段,其目的就是通过对已知数据的学习,对未知的数据进行预测.第一章中主要介绍了一些基本的概念如监督学习\非监督学习,损失函数的定义,过拟合与欠拟合,回归和分类任务等基本的概念在此我就不细说了,下面重点]]></description><category></category></item><item><title><![CDATA[移植较新（Linux3.19）内核至mini2440开发板（二）——LCD的移植]]></title><link>https://blog.youkuaiyun.com/winter_evening/article/details/45290555</link><guid>https://blog.youkuaiyun.com/winter_evening/article/details/45290555</guid><author>winter_evening</author><pubDate>Sun, 26 Apr 2015 18:32:32 +0800</pubDate><description><![CDATA[mini2440 LCD的移植]]></description><category></category></item><item><title><![CDATA[移植较新（Linux3.19）内核至mini2440开发板（一）]]></title><link>https://blog.youkuaiyun.com/winter_evening/article/details/45254437</link><guid>https://blog.youkuaiyun.com/winter_evening/article/details/45254437</guid><author>winter_evening</author><pubDate>Fri, 24 Apr 2015 23:13:05 +0800</pubDate><description><![CDATA[最新内核移植]]></description><category></category></item><item><title><![CDATA[关于mini2440挂载nfs文件系统一些错误的解决方法]]></title><link>https://blog.youkuaiyun.com/winter_evening/article/details/45156839</link><guid>https://blog.youkuaiyun.com/winter_evening/article/details/45156839</guid><author>winter_evening</author><pubDate>Mon, 20 Apr 2015 22:22:05 +0800</pubDate><description><![CDATA[在挂载nfs根文件系统时遇到了两个问题，第一个是内核启动到最后出现如下信息：


VFS: Mounted root (nfs filesystem).

Freeing init memory: 144K

Warning: unable to open an initial console.
这个问题足足困扰了我两天的时间，刚开始一直怀疑是nfs服务器配置的问题，尝试了很久都无]]></description><category></category></item><item><title><![CDATA[matlab在建模中的应用]]></title><link>https://blog.youkuaiyun.com/winter_evening/article/details/43154727</link><guid>https://blog.youkuaiyun.com/winter_evening/article/details/43154727</guid><author>winter_evening</author><pubDate>Mon, 26 Jan 2015 17:10:34 +0800</pubDate><description><![CDATA[1.polyfit(X,Y,N)多项式拟合，返回降幂排列的多项式
polyval(P,xi)计算多项式的值


2.]]></description><category></category></item><item><title><![CDATA[Lingo规划问题求解]]></title><link>https://blog.youkuaiyun.com/winter_evening/article/details/43054359</link><guid>https://blog.youkuaiyun.com/winter_evening/article/details/43054359</guid><author>winter_evening</author><pubDate>Fri, 23 Jan 2015 13:21:43 +0800</pubDate><description><![CDATA[格式


model:
max=2*x1+3*x2+4*x3;
1.5*x1+3*x2+5*x3
280*x1+250*x2+400*x3
@gin(x1);@gin(x2);@gin(x3); !如果取x1,x2,x3为整数
end]]></description><category></category></item><item><title><![CDATA[MATLAB绘图]]></title><link>https://blog.youkuaiyun.com/winter_evening/article/details/43031195</link><guid>https://blog.youkuaiyun.com/winter_evening/article/details/43031195</guid><author>winter_evening</author><pubDate>Thu, 22 Jan 2015 22:52:11 +0800</pubDate><description><![CDATA[1.fplot可用来精确绘图
例如
sn = @(x) sin(1./x);
fplot(sn,[0.01,0.1])



或者


Create a file named myfun.m that containsthe following code.
function Y = myfun(x)
Y(:,1) = 200*sin(x(:))./x(:);
Y(:,2)]]></description><category></category></item><item><title><![CDATA[使用MATLAB进行一次和二次规划问题求解]]></title><link>https://blog.youkuaiyun.com/winter_evening/article/details/42985613</link><guid>https://blog.youkuaiyun.com/winter_evening/article/details/42985613</guid><author>winter_evening</author><pubDate>Wed, 21 Jan 2015 22:59:59 +0800</pubDate><description><![CDATA[对于一次问题，可以使用linprog函数，例如
Find x that minimizes 


f(x) = –5x1 –4x2 –6x3,


subject to


x1 – x2 + x3 ≤20
3x1 +2x2 + 4x3 ≤42
3x1 +2x2 ≤ 30
0≤ x1, 0 ≤ x2,0 ≤ x3.


First, enter the coeffi]]></description><category></category></item></channel></rss>