How to add a key manually in PySpark

Question

I have a scala code in PySpark.

      val rdd =sparkContext.parallelize(( (1 to 20).map(x=>("key",x))), 4)
      rdd.reduceByKey(_ + _)
      rdd