`
coderplay
  • 浏览: 571841 次
  • 性别: Icon_minigender_1
  • 来自: 广州杭州
社区版块
存档分类
最新评论
文章列表
好记性不如烂笔头,记一下: NetflixDataPrep(准备数据) -> NetflixCanopyMaker(产生canopy中心) -> NetflixCanopyData(分配所有点到各canopy) -> NetflixKMeansIter(进行k-means聚类) 假设数据记录条数为n, 第二步产生的canopy数量为c, 那第三步计算量则为 n * c,就算用了mapper计算量也非常大。一定要改进成增量式的方法。 研究一下~~

asguard

interface使用Multiname,如果a.swf
k-Means, k-Medoids, Kernel Clustering, Spectral Clustering (uses eigenvectors), Gravitational Clustering, Canopy Clustering, Self-Organizing Maps, Expectation Maximization, AGNES, CLARA, DBSCAN, DIANA, BIRCH, and many others.
Today, I accidently found an interesting stuff, which may help us operating large scales of data sets for redpoll. This is a matrix computational library based on hadoop hbase. http://code.google.com/p/hama/
Days before, I've submitted an application to participate in Apache Mahout and  at this time, have got a reply from the guru of this project. It inspired us with lots of courages. We decided that if I were selected by ASF, we will integrate redpoll into Mahout which has the same end goals, same licen ...
Personal Information Name:周敏 (Min Zhou) Gender:male Date of Birth:1982/04 Current WorkState:Only the good opportunity then i consider Primary Contact:Mobile (13560220768)            Contact2:School(020-38375371) Web Page:http://coderplay.iteye.com Wed State:Single Political Background:Non-Party M ...
We are pleased to introduce a new open source proejct today. It's another machine learning library using hadoop besides the mahout of ASF(Apache Software Foundation). The name of this project is redpoll, which means any of several small finches of northern  North America and Eruasia, having a red crw ...

texlive使用日记

    博客分类:
  • misc
往代码里面插入中文。 这段有时候会出错,编译不了。 但大多时候可以正常运行。 \lstset{escapebegin=\textbraceleft\begin{CJK}{GB}{gbsn},escapeend=\end{CJK}\textbraceright} \begin{lstlisting}[escapeinside=\{\}] for i:=maxint to 0 do begin { 中文 } end; \end{lstlisting} 以下都可以改中文 \renewcommand\abstractname{摘~~要} \renewcommand{\contentsname}{目 ...
\input与\include有区别. 比如写Makefile的时候, Makefile所在目录是项目的主目录. 此项目下面有src/chapter1.tex, src/main.tex 如果main.tex用\input引用chapter1.tex时,得带上目录名 \input{src/chapter1} 表明此时目录是和Makefile的地址一致。 而\include却不用,它是 \include{chapter1} 2. \renewcommand{}{}可以用来重定义一些命令的相关参数。同类型的还有\newenvironment{}{}{},它是定义环境的,更大幅度地改变。 \re ...
安装 1.  在http://www.tug.org/texlive/下载texlive的iso文件. 它分live和dist版, windows上可以用dist版. live版内容太多. 2. 安装好了后, 会有一个texmf-local目录与texlive的根目录平行. 这个texmf-local目录也就是你自己的宏包,字体等资源的添加目录. 3. 安装中文字体. 用王垠的gbkfonts来做字体文件.具体方法参见: http://learn.tsinghua.edu.cn:8080/2001315450/tex/tex_setup_chinese.html windows上可以下载 ...
<!-- @page { size: 21cm 29.7cm; margin: 2cm } P { margin-bottom: 0.21cm } --> /usr/lib/jvm/java-6-sun/bin/java -Xmx1000m   -Dhadoop.log.dir=/home/hadoop01/hadoop/hadoop-0.14.3/logs -Dhadoop.log.file=hadoop.log -Dhadoop.home.dir=/home/hadoop01/hadoop/hadoop-0.14.3 ...
新的ubuntu系统会缺少一些工具 和lib. 用apt-get来获得 apt-get install build-essential apt-get install libncurses5-dev apt-get install m4 apt-get install libssl-dev 接着再编译就okay了. 运行 # erl 1> appmon:start(). 会发现界面没有, 程序停在那里. 这是因为缺少了tcl/tk库.  apt-get install  tk8.3-dev 装的时候, 会出现如下问题: # erl Erlang (BEAM) emulator ...
java 代码 //MethodLinker.java里面 // 这个函数实现了MemberVisitor的接口. public void visitAnyMember(Clazz clazz, Member member) { // 取得类成员的名称和描述符 String name = member.getName(clazz); String descriptor = member.getDescriptor(clazz); // 特殊情况: 如果是初始化或构 ...
--------- driver篇 --------------- driver一般通过erl_ddll.erl中的load_driver/2函数来实现. load_driver(Path, Driver) ->     do_load_driver(Path, Driver, [{driver_options,[kill_ports]}]). ... do_load_driver(Path, Driver, DriverFlags) ->     case erl_ddll:try_load(Path, Driver,[{monitor,pending_driver}]++Dri ...
    也许很多人碰上过, 用example1_lid:start()会碰上如下错误:      ** exited: {error,could_not_load_driver} **     用gdb跟了一下, 发现其实load driver是没有问题, 问题出在load driver之后的版本匹配, driver_incorrect_version.     解决办法很简单,两步:   1. 把example1_lid.c中的    #include "erl_driver.h" 改成   #include <<erl_driver.h>erl_dr ...
Global site tag (gtag.js) - Google Analytics