I have to try but I think it probably will happen with spark-shell as well.
I have found alternative to pass a Array to UDF as a parameter.
Thanks
On Thu, Nov 17, 2016 at 9:24 AM, moon soo Lee wrote:
> Are you able to run the same code in SPARK_HOME/bin/spark-shell ?
>
> Thanks,
> moon
>
> On M
Are you able to run the same code in SPARK_HOME/bin/spark-shell ?
Thanks,
moon
On Mon, Nov 14, 2016 at 2:21 PM Nirav Patel wrote:
> I have a following map:
>
> final val idxMapArr = idxMap.collectAsMap
>
> Which is being used in one of spark transformation here:
>
> def labelStr(predictions: Wr
I have a following map:
final val idxMapArr = idxMap.collectAsMap
Which is being used in one of spark transformation here:
def labelStr(predictions: WrappedArray[Double]): Array[String] = {
predictions.array.map(*idxMapArr*.getOrElse(_, "Other"))
}
I get following error :
Caused by: java
