org.deeplearning4j.spark.impl.common.CountPartitionsFunction Maven / Gradle / Ivy
package org.deeplearning4j.spark.impl.common;
import org.apache.spark.api.java.JavaRDD;
import org.apache.spark.api.java.function.Function2;
import org.deeplearning4j.spark.api.Repartition;
import scala.Tuple2;
import java.util.Collections;
import java.util.Iterator;
/**
* This is a function use to count the number of elements in each partition.
* It is used as part of {@link org.deeplearning4j.spark.util.SparkUtils#repartitionBalanceIfRequired(JavaRDD, Repartition, int, int)}
*
* @author Alex Black
*/
public class CountPartitionsFunction implements Function2, Iterator>> {
@Override
public Iterator> call(Integer v1, Iterator v2) throws Exception {
int count = 0;
while(v2.hasNext()){
v2.next();
count++;
}
return Collections.singletonList(new Tuple2<>(v1, count)).iterator();
}
}