本文整理汇总了Java中org.apache.spark.api.java.JavaPairRDD.mapValues方法的典型用法代码示例。如果您正苦于以下问题：Java JavaPairRDD.mapValues方法的具体用法？Java JavaPairRDD.mapValues怎么用？Java JavaPairRDD.mapValues使用的例子？那么恭喜您, 这里精选的方法代码示例或许可以为您提供帮助。您也可以进一步了解该方法所在类org.apache.spark.api.java.JavaPairRDD的用法示例。
在下文中一共展示了JavaPairRDD.mapValues方法的2个代码示例，这些例子默认根据受欢迎程度排序。您可以为喜欢或者感觉有用的代码点赞，您的评价将有助于系统推荐出更棒的Java代码示例。
示例1: getSupportOfRelations
import org.apache.spark.api.java.JavaPairRDD; //导入方法依赖的package包/类
public JavaPairRDD<String,Float> getSupportOfRelations(JavaPairRDD<String,List<Tuple2<Integer, Integer>>> relationIndex, long numEntititiesSquared, float minSupportThreshold) {
JavaPairRDD<String, Float> unnormalizedSupports = relationIndex
.mapValues(so -> (float)so.size() / numEntititiesSquared);
unnormalizedSupports.setName("unnormalizedSupports").cache();
System.out.println(unnormalizedSupports.count()+" relations have been assigned a support value"); // dummy action to trigger execution
float max_support = unnormalizedSupports.values().max(Ordering.natural());
return unnormalizedSupports
.mapValues(x-> x/max_support) //normalize the support values
.filter(x-> x._2()> minSupportThreshold); //filter out relations below the min support threshold (infrequent relations)
}
示例2: getDiscriminabilityOfRelations
import org.apache.spark.api.java.JavaPairRDD; //导入方法依赖的package包/类
public JavaPairRDD<String,Float> getDiscriminabilityOfRelations(JavaPairRDD<String,List<Tuple2<Integer, Integer>>> relationIndex) {
return relationIndex.mapValues(soIterable -> {
int frequencyOfRelation = 0;
IntOpenHashSet localObjects = new IntOpenHashSet();
for (Tuple2<Integer, Integer> so : soIterable) {
frequencyOfRelation++;
localObjects.add(so._2());
}
return (float) localObjects.size() / frequencyOfRelation;
});
}