Spark中组件Mllib的学习19之分层抽样

更多代码请见：https://github.com/xubo245/SparkLearning
Spark中组件Mllib的学习之基础概念篇
1解释
分层抽样的概念就不讲了，具体的操作：
RDD有个操作可以直接进行抽样：sampleByKey和sample等，这里主要介绍这两个
（1）将字符串长度为2划分为层1和层2，对层1和层2按不同的概率进行抽样
数据

aa
bb
cc
dd
ee
aaa
bbb
ccc
ddd
eee

比如：
val fractions: Map[Int, Double] = (List((1, 0.2), (2, 0.8))).toMap //设定抽样格式
sampleByKey(withReplacement = false, fractions, 0)
fractions表示在层1抽0.2，在层2中抽0.8
withReplacement false表示不重复抽样
0表示随机的seed

源码：

  /*** Return a subset of this RDD sampled by key (via stratified sampling).** Create a sample of this RDD using variable sampling rates for different keys as specified by* `fractions`, a key to sampling rate map, via simple random sampling with one pass over the* RDD, to produce a sample of size that's approximately equal to the sum of* math.ceil(numItems * samplingRate) over all key values.** @param withReplacement whether to sample with or without replacement* @param fractions map of specific keys to sampling rates* @param seed seed for the random number generator* @return RDD containing the sampled subset*/def sampleByKey(withReplacement: Boolean,fractions: Map[K, Double],seed: Long = Utils.random.nextLong): RDD[(K, V)] = self.withScope {require(fractions.values.forall(v => v >= 0.0), "Negative sampling rates.")val samplingFunc = if (withReplacement) {StratifiedSamplingUtils.getPoissonSamplingFunction(self, fractions, false, seed)} else {StratifiedSamplingUtils.getBernoulliSamplingFunction(self, fractions, false, seed)}self.mapPartitionsWithIndex(samplingFunc, preservesPartitioning = true)}

（2）和（3）类似，请看代码

2.代码：

/*** @author xubo*         ref:Spark MlLib机器学习实战*         more code:https://github.com/xubo245/SparkLearning*         more blog:http://blog.csdn.net/xubo245*/
package org.apache.spark.mllib.learning.basicimport org.apache.spark.{SparkConf, SparkContext}/*** Created by xubo on 2016/5/23.*/
object StratifiedSamplingLearning {def main(args: Array[String]) {val conf = new SparkConf().setMaster("local[4]").setAppName(this.getClass().getSimpleName().filter(!_.equals('$')))val sc = new SparkContext(conf)println("First:")val data = sc.textFile("file/data/mllib/input/basic/StratifiedSampling.txt") //读取数.map(row => {//开始处理if (row.length == 3) //判断字符数(row, 1) //建立对应mapelse (row, 2) //建立对应map}).map(each => (each._2, each._1))data.foreach(println)println("sampleByKey:")val fractions: Map[Int, Double] = (List((1, 0.2), (2, 0.8))).toMap //设定抽样格式val approxSample = data.sampleByKey(withReplacement = false, fractions, 0) //计算抽样样本approxSample.foreach(println)println("Second:")//http://homepage.cs.latrobe.edu.au/zhe/ZhenHeSparkRDDAPIExamples.html#sampleByKeyval randRDD = sc.parallelize(List((7, "cat"), (6, "mouse"), (7, "cup"), (6, "book"), (7, "tv"), (6, "screen"), (7, "heater")))val sampleMap = List((7, 0.4), (6, 0.8)).toMapval sample2 = randRDD.sampleByKey(false, sampleMap, 42).collectsample2.foreach(println)println("Third:")//http://bbs.csdn.net/topics/390953396val a = sc.parallelize(1 to 20, 3)val b = a.sample(true, 0.8, 0)val c = a.sample(false, 0.8, 0)println("RDD a : " + a.collect().mkString(" , "))println("RDD b : " + b.collect().mkString(" , "))println("RDD c : " + c.collect().mkString(" , "))sc.stop}
}

3.结果：

First:
2016-05-23 22:37:34 WARN  :139 - Your hostname, xubo-PC resolves to a loopback/non-reachable address: fe80:0:0:0:482:722f:5976:ce1f%20, but we couldn't find any external IP address!
(2,aa)
(1,bbb)
(2,bb)
(1,ccc)
(2,cc)
(1,ddd)
(2,dd)
(1,eee)
(2,ee)
(1,aaa)
sampleByKey:
(2,aa)
(2,bb)
(2,cc)
(2,ee)
Second:
(7,cat)
(6,mouse)
(6,book)
(6,screen)
(7,heater)
Third:
RDD a : 1 , 2 , 3 , 4 , 5 , 6 , 7 , 8 , 9 , 10 , 11 , 12 , 13 , 14 , 15 , 16 , 17 , 18 , 19 , 20
RDD b : 2 , 4 , 5 , 6 , 10 , 14 , 19 , 20
RDD c : 1 , 2 , 4 , 5 , 8 , 10 , 12 , 13 , 14 , 15 , 16 , 17 , 18 , 19 , 20

参考
【1】http://spark.apache.org/docs/1.5.2/mllib-guide.html
【2】http://spark.apache.org/docs/1.5.2/programming-guide.html
【3】https://github.com/xubo245/SparkLearning

Spark中组件Mllib的学习19之分层抽样相关推荐

Spark中组件Mllib的学习16之分布式行矩阵的四种形式
来源:http://blog.csdn.net/xubo245/article/details/51483995 更多代码请见:https://github.com/xubo245/SparkLear ...
Spark中组件Mllib的学习15之创建分布式矩阵
更多代码请见:https://github.com/xubo245/SparkLearning Spark中组件Mllib的学习之基础概念篇 1解释创建分布式矩阵 2.代码: /*** @autho ...
Spark中组件Mllib的学习40之梯度提升树（GBT）用于回归
更多代码请见:https://github.com/xubo245/SparkLearning Spark中组件Mllib的学习之分类篇 1解释 GBRT(Gradient Boost Regr ...
Spark中组件Mllib的学习41之保序回归（Isotonic regression）
更多代码请见:https://github.com/xubo245/SparkLearning Spark中组件Mllib的学习之分类篇 1解释问题描述:给定一个无序数字序列,要求不改变每个元素的位 ...
Spark中组件Mllib的学习11之使用ALS对movieLens中一百万条（1M）数据集进行训练，并对输入的新用户数据进行电影推荐
更多代码请见:https://github.com/xubo245/SparkLearning 1解释 spark-1.5.2 数据集:http://grouplens.org/datasets/mo ...
Spark中组件Mllib的学习1之Kmeans错误解决
更多代码请见:https://github.com/xubo245/SparkLearning 解决办法:(中间比较多,为了方便看到,放在最开始) txt文件格式不对,用WPS转存的是UTF-16,s ...
Spark四大组件包括Spark Streaming、Spark SQL、Spark MLlib和Spark GraphX。
Spark四大组件包括Spark Streaming.Spark SQL.Spark MLlib和Spark GraphX.它们的主要应用场景是: Spark Streaming: Spark Str ...
中国Spark技术峰会（上）：Spark与生态圈中组件结合实战
5月13日-15日,由全球最大中文IT社区CSDN主办的"2016中国云计算技术大会"(Cloud Computing Technology Conference 2016,简称C ...
jsx 调用php,JavaScript_JavaScript的React框架中的JSX语法学习入门教程，什么是JSX？在用React写组件的 - phpStudy...
JavaScript的React框架中的JSX语法学习入门教程什么是JSX? 在用React写组件的时候,通常会用到JSX语法,粗看上去,像是在Javascript代码里直接写起了XML标签,实质上 ...

Spark中组件Mllib的学习19之分层抽样

Spark中组件Mllib的学习19之分层抽样相关推荐

最新文章

热门文章