Hello Aljoscha,
For *DataSet: *
IterativeDataSet *loop* = centroids.iterate(numIterations);
DataSet *newCentroids* = points.map(new SelectNearestCenter()).
*withBroadcastSet*(*loop*, "*centroids*")
.map(new CountAppender()).groupBy(0).reduce(new CentroidAccumulator())
.map(new CentroidAverager());
There you have your explanation. A loop actually has to be a loop for it to
work in Flink.
On Sat, 14 May 2016 at 16:35 subash basnet wrote:
> Hello,
>
> I had to use,
> private static IterativeStream *loop*;
> loop as global variable because I cannot broadcast it like that of DataSet
> API in D
Hello,
I had to use,
private static IterativeStream *loop*;
loop as global variable because I cannot broadcast it like that of DataSet
API in DataStream API.
I tried to use *closewith * like that of DataSet as below in DataStream but
it gives me exception:
DataStream finalCentroids = *loop*.close
Hello Aljoscha,
Below is the shorted version of StockAnalysis class which is a datastream
adapation of the *KMeans.java* dataset code.
public class StockAnalysis{
public static void main(String args[]){
DataStream centroids = newCentroidDataStream.map(new
TupleCentroidConverter());
*
Could you please post your code.
On Sat, 7 May 2016 at 19:16 subash basnet wrote:
> Hello all,
>
> I am getting the below error on execute of StreamExecutionEnvironment.
>
>
> *Caused by: java.lang.IllegalStateException: Iteration
> FeedbackTransformation{id=15, name='Feedback',
> outputType=Poj
Hello all,
I am getting the below error on execute of StreamExecutionEnvironment.
*Caused by: java.lang.IllegalStateException: Iteration
FeedbackTransformation{id=15, name='Feedback',
outputType=PojoType]>, parallelism=4} does not have any feedback
edges.*
The run method inside the thread class
