jyp
@proflin 请问spark stream支持几种window operation？ 在论文上看到一种叫做session window的
话说前两种window operation比较适合的场景是什么？ 我在一些论文上看到watermark这个概念，据说用这个概念可以取得correctness ， latency的一种平衡。 对了，我觉得像交易所这种数据对corretness的要求是极其高的，我一直觉得不适合用现在这些流失处理进行，感觉还是batch处理比较合理。 @proflin
@twtygqyy I didn't find any info about 0-1 normalized in the code.... :-)
@twtygqyy another question here, the size of the generated .h5 file for 291 png(13M) is about 14G. If I want to use 800 pngs(1000x800) to generate .h5 file => It...
@debasishg cannot pass again. The same problem still exist. my scala version is 2.10.6 here and java version is "1.7.0_80".
@debasishg The ++ 2.10.6 command has no help to the problem ｓｏｌｖｉｎｇ. And the follwing is the content of my build.sbt: ``` libraryDependencies ++= Seq( "net.debasishg" %% "redisclient" % "3.2"...
I have past the compile dependency in the built.sbt, but failed in sbt test due to ``` [info] - should distribute work amongst the clients for 2000000 list operations ***...