148、Spark核心编程进阶之takeSampled以及公司年
2019-01-22 本文已影响0人
ZFH__ZJ
takeSample算子,与sample不同之处,两点
- action操作,sample是transformation操作
- 不能指定抽取比例,只能是抽取几个
public class TakeSampled {
public static void main(String[] args) {
SparkConf conf = new SparkConf()
.setAppName("TakeSampled")
.setMaster("local");
JavaSparkContext sc = new JavaSparkContext(conf);
List<String> staffList = Arrays.asList("张三", "李四", "王二", "麻子",
"赵六", "王五", "李大个", "王大妞", "小明", "小倩");
JavaRDD<String> staffRDD = sc.parallelize(staffList);
List<String> luckyStaffList = staffRDD.takeSample(false, 3);
for(String luckyStaff : luckyStaffList) {
System.out.println(luckyStaff);
}
sc.close();
}
}