python 安装pyspark_python pyspark入门篇

一.环境介绍：

1.安装jdk 7以上

2.python 2.7.11

3.IDE pycharm

4.package: spark-1.6.0-bin-hadoop2.6.tar.gz

二.Setup

1.解压spark-1.6.0-bin-hadoop2.6.tar.gz 到目录D:\spark-1.6.0-bin-hadoop2.6

2.配置环境变量Path，添加D:\spark-1.6.0-bin-hadoop2.6\bin,此后可以在cmd端输入pySpark,返回如下则安装完成：

3.将D:\spark-1.6.0-bin-hadoop2.6\python下的pySpark文件拷贝到C:\Python27\Lib\site-packages

4.安装py4j ， pip install py4j -i https://pypi.douban.com/simple

5.配置pychar环境变量：

三.Example

1.make a new python file: wordCount.py

#!/usr/bin/env python#-*- coding: utf-8 -*-

importsysfrom pyspark importSparkContextfrom operator importaddimportredefmain():

sc= SparkContext(appName= "wordsCount")

lines= sc.textFile(‘words.txt‘)

counts= lines.flatMap(lambda x: x.split(‘ ‘)) .map(lambda x : (x, 1)) .reduceByKey(add)

output=counts.collect()printoutputfor (word, count) inoutput:print "%s: %i" %(word, count)

sc.stop()if __name__ =="__main__":

main()

2.代码中的words.txt如下：

The dynamic lifestyle

people lead nowadays

causes many reactionsin our bodies andthe one thatisthe

most frequent of allis the headache

3.配置pycharm环境变量：

3.1 工具栏 run --> Edit configuration-->点击箭头位置

3.2  然后点击 + ，输入key：SPARK_HOME， value： D:\spark-1.6.0-bin-hadoop2.6

4.输出结果如下图：

四.深入练习：

1.文档：http://spark.apache.org/docs/latest/api/python/pyspark.html

2.在解压的Spark文档下，有example下有很多实例可以练习。D:\spark-1.6.0-bin-hadoop2.6\examples\src\main\python