pyspark.streaming.DStream.combineByKey

DStream.combineByKey(createCombiner: Callable[[V], U], mergeValue: Callable[[U, V], U], mergeCombiners: Callable[[U, U], U], numPartitions: Optional[int] = None) → pyspark.streaming.dstream.DStream[Tuple[K, U]][source]

Return a new DStream by applying combineByKey to each RDD.