String map(Tuple2 tuple) throws Exception {
System.out.println(tuple);
return null;
}
@Override
public void onEpochWatermarkIncremented(int i, Context context,
Collector collector) throws Exception {
System.out.println("i");
"statedescriptor", // the state
> name
> BasicTypeInfo.DOUBLE_TYPE_INFO,
> BasicTypeInfo.STRING_TYPE_INFO
> );
>
DataStreamList.of(newModelUpdate),
DataStreamList.of(finalOutput));
});
result.get(0).print();
// Execute program
env.execute("Flink Java API Skeleton");
}
}
Best,
Komal
From: Yunfeng Zhou
Date
Hi Komal,
For the first question, could you please provide a simple program that
could help reproduce this exception? That could help us better find
out the bugs (if any) in Flink ML.
For the second question, there have been Functions implementing the
IterationListener interface in Flink ML[1] an
Hi Flink Dev Team,
I have two possible bugs to report for Flink ML Iteration.
Flink v1.17.2
Flink ML v2.3.0
Java 11
Bug # 1
Implementing a UDF KeyedRichCoProcessFunction or CoFlatMapFunction inside
IterationBody yields a “java.lang.ClassCastException:
org.apache.flink.iteration.IterationRecord c
