<?xml version="1.0" encoding="utf-8" ?><rss version="2.0"><channel><title><![CDATA[LULIN60的专栏]]></title><description><![CDATA[]]></description><link>https://blog.youkuaiyun.com/LULIN60</link><language>zh-cn</language><generator>https://blog.youkuaiyun.com/</generator><copyright><![CDATA[Copyright &copy; LULIN60]]></copyright><item><title><![CDATA[二维坐标点中查找最大凸多边形Java实现]]></title><link>https://blog.youkuaiyun.com/LULIN60/article/details/54615991</link><guid>https://blog.youkuaiyun.com/LULIN60/article/details/54615991</guid><author>LULIN60</author><pubDate>Thu, 19 Jan 2017 14:42:39 +0800</pubDate><description><![CDATA[参考如下文献中的Ruby代码实现的：
http://www.jianshu.com/p/a297324f4906]]></description><category></category></item><item><title><![CDATA[Spark MLlib中FPGrowth和FPTree详解之二]]></title><link>https://blog.youkuaiyun.com/LULIN60/article/details/52263388</link><guid>https://blog.youkuaiyun.com/LULIN60/article/details/52263388</guid><author>LULIN60</author><pubDate>Sat, 20 Aug 2016 20:35:49 +0800</pubDate><description><![CDATA[这一章节，主要介绍FPGrowth源码，以及运行过程演示
2.3 FPGrowth源码详解

run方法是FPGrowth的入口函数，其代码注释如下：
/**
  * Computes an FP-Growth model that contains frequent
itemsets.
  * @param data input data set, each element con]]></description><category></category></item><item><title><![CDATA[Spark MLlib中FPGrowth和FPTree详解之一]]></title><link>https://blog.youkuaiyun.com/LULIN60/article/details/52255242</link><guid>https://blog.youkuaiyun.com/LULIN60/article/details/52255242</guid><author>LULIN60</author><pubDate>Sat, 20 Aug 2016 00:31:42 +0800</pubDate><description><![CDATA[一、准备知识
1.1 Scala版本：2.10.4
1.2 Spark版本：1.5.0
Spark中实现关联规则算法的包是：org.apache.spark.mllib.fpm。包中的文件如下图所示：



这里面我重点讲解红色箭头指向的两个代码文件。讲解过程中如果有误解的地方，还请评论指正，谢谢！
1.3 频繁模式增长FP-Growth
要理解Spark MLlib中FPGro]]></description><category></category></item><item><title><![CDATA[IntelliJ IDEA and Scala tutorial]]></title><link>https://blog.youkuaiyun.com/LULIN60/article/details/52237509</link><guid>https://blog.youkuaiyun.com/LULIN60/article/details/52237509</guid><author>LULIN60</author><pubDate>Thu, 18 Aug 2016 10:06:52 +0800</pubDate><description><![CDATA[下面是完整的利用IntelliJ进行Scala开发的教程，摘自Cousera的Scala公开课，网址如下：


参考资料
http://www.scala-sbt.org/release/docs/Setup.html



第一步：要安装JDK，程序猿都知道！


Installing the JDK

Linux

Ubuntu, Debian: To insta]]></description><category></category></item><item><title><![CDATA[Apache Hadoop 安装]]></title><link>https://blog.youkuaiyun.com/LULIN60/article/details/50429526</link><guid>https://blog.youkuaiyun.com/LULIN60/article/details/50429526</guid><author>LULIN60</author><pubDate>Tue, 29 Dec 2015 21:40:08 +0800</pubDate><description><![CDATA[Apache Hadoop 安装过程
tar -zxvf hadoop-2.7.0 -C ...
其中关键的文件夹为 bin sbin 和 etc
一、修改配置文件
1.1 vim Hadoop-env.sh
$JAVA_HOME的路径拷贝到export JAVA_HOME=...
1.2 vim core-site.xml


fs.defaultFS
hdfs://`hos]]></description><category></category></item><item><title><![CDATA[分类算法精华理解]]></title><link>https://blog.youkuaiyun.com/LULIN60/article/details/50421091</link><guid>https://blog.youkuaiyun.com/LULIN60/article/details/50421091</guid><author>LULIN60</author><pubDate>Mon, 28 Dec 2015 19:58:44 +0800</pubDate><description><![CDATA[一、决策树Decision Tree
1.1 抛开那些复杂的理论，首先需要牢记的就是信息熵公式：Information entropy = -plog2(p).
1.2 ID3算法的信息增益
1.2.1  利用决策列（就是训练集中已经知道的类别那一列来计算整体信息熵，以两个类别为例C1, C2)对所有数据进行划分。假如n个训练集中有n1个子集属于同类，另外n2属于另一个类。那么p1=n1/n]]></description><category></category></item><item><title><![CDATA[Win7中Eclipse开发运行Hadoop程序的相关配置]]></title><link>https://blog.youkuaiyun.com/LULIN60/article/details/50384213</link><guid>https://blog.youkuaiyun.com/LULIN60/article/details/50384213</guid><author>LULIN60</author><pubDate>Wed, 23 Dec 2015 08:47:34 +0800</pubDate><description><![CDATA[首先需要强调的是，Hadoop集群还是只能在linux系统中进行部署，我用的CentOS6.4版本的系统，安装的是Hadoop2.6.0.

环境：
win7 64位  通过命令winver查看系统版本
hadoop-2.6.0  通过Hadoop -v 来查看
Eclipse  Version: Mars.1 Release (4.5.1) Build id: 20150924-120]]></description><category></category></item><item><title><![CDATA[通过PATH-B：cloudera manager安装cloudera5.4]]></title><link>https://blog.youkuaiyun.com/LULIN60/article/details/49868577</link><guid>https://blog.youkuaiyun.com/LULIN60/article/details/49868577</guid><author>LULIN60</author><pubDate>Mon, 16 Nov 2015 17:04:30 +0800</pubDate><description><![CDATA[首先，最好的安装指导还是官网的安装手册，最清晰的安装步骤就是手册里面的图。然后我就是安装通过PATH B来进行手动安装的。
首先列出系统和安装包：
虚拟机：VMware
操作系统：CentOS-6.5-x86_64-bin-DVD1.iso （用这个系统在安装agent的时候需要三个RPM包：init-functions,gassapi和portmap，按照提示通过yum安装就好了）
软件]]></description><category></category></item></channel></rss>