0

I'm getting below error while running spark program using spark-submit.

My spark-cluster is of version 2.0.0 and I use sbt to compile my code and below is my sbt dependencies. 

libraryDependencies ++= Seq(
  "commons-io" % "commons-io" % "2.4",
  "com.google.guava" % "guava" % "19.0",
  "jfree" % "jfreechart" % "1.0.13",
  ("org.deeplearning4j" % "deeplearning4j-core" % "0.5.0").exclude("org.slf4j", "slf4j-log4j12"),
  "org.jblas" % "jblas" % "1.2.4",
  "org.nd4j" % "canova-nd4j-codec" % "0.0.0.15",
  "org.nd4j" % "nd4j-native" % "0.5.0" classifier "" classifier "linux-x86_64",
  "org.deeplearning4j" % "dl4j-spark" % "0.4-rc3.6" ,
  "org.apache.spark" % "spark-sql_2.10" % "1.3.1", 
  "org.apache.spark" % "spark-hive_2.10" % "1.3.1",
  "org.apache.hive" % "hive-serde" % "0.14.0", 
  ("org.deeplearning4j" % "arbiter-deeplearning4j" % "0.5.0"))



16/11/14 22:57:03 INFO hive.HiveSharedState: Warehouse path is 'file:/home/hduser/spark-warehouse'.
Exception in thread "main" java.lang.NoSuchMethodError: org.apache.spark.sql.hive.HiveContext.sql(Ljava/lang/String;)Lorg/apache/spark/sql/DataFrame;
    at poc.common.utilities.StockData$.fetchStockData(StockData.scala:15)
    at poc.analaticsEngine.AnalaticsStockWorkBench.fetchTrainingDataSet(AnalaticsStockWorkBench.scala:69)
    at poc.analaticsEngine.AnalaticsStockWorkBench.trainModel(AnalaticsStockWorkBench.scala:79)
    at test.poc.analatics.StockPrediction$.testTrainSaveModel(StockPrediction.scala:21)
    at test.poc.analatics.StockPrediction$.main(StockPrediction.scala:10)
    at test.poc.analatics.StockPrediction.main(StockPrediction.scala)
    at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
    at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:62)
    at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43)
    at java.lang.reflect.Method.invoke(Method.java:497)
    at org.apache.spark.deploy.SparkSubmit$.org$apache$spark$deploy$SparkSubmit$$runMain(SparkSubmit.scala:729)
    at org.apache.spark.deploy.SparkSubmit$.doRunMain$1(SparkSubmit.scala:185)
    at org.apache.spark.deploy.SparkSubmit$.submit(SparkSubmit.scala:210)
    at org.apache.spark.deploy.SparkSubmit$.main(SparkSubmit.scala:124)
    at org.apache.spark.deploy.SparkSubmit.main(SparkSubmit.scala)
16/11/14 22:57:03 INFO spark.SparkContext: Invoking stop() from shutdown hook
Charls Joseph
  • 141
  • 2
  • 9
  • The point where it is failed is below and is working locally in spark-local val sqlData = Spark.hiveContext.sql(hiveQueryForStockCloseValue); – Charls Joseph Nov 14 '16 at 09:30

2 Answers2

0

First of all, you're saying that you use Spark 2.0.0, but in your dependencies you have 

"org.apache.spark" % "spark-sql_2.10" % "1.3.1", 
"org.apache.spark" % "spark-hive_2.10" % "1.3.1",

You need to change those dependencies to version 2.0.0 to keep it consistent with Spark. What is more, you don't need to specify spark-sql dependency separately, because it's already contained in spark-hive. hive-serde also comes in 2.1.0 version now, so 0.14 is probably obsolete.

Shem
  • 565
  • 2
  • 10
  • Yea. 2.0.0 version update is on my mind. But 1.3.1 was already added into my jar and it shouldn't complain about the methodnotfound is what I', thinking. – Charls Joseph Nov 14 '16 at 09:53
0

Please follow the dl4j examples for versioning. I'm not sure where or how you got canova in there (we haven't used it for nearly 6 months now?)

https://github.com/deeplearning4j/dl4j-examples/blob/master/dl4j-examples/pom.xml

Adam Gibson
  • 3,055
  • 1
  • 10
  • 12
  • canova was added mistakenly and not being used. – Charls Joseph Nov 15 '16 at 08:43
  • I'm getting another error :Exception in thread "main" java.lang.NoSuchMethodError: org.apache.spark.Accumulator.(Ljava/lang/Object;Lorg/apache/spark/AccumulatorParam;)V at org.deeplearning4j.spark.impl.common.SumAccum.(SumAccum.java:14) at org.deeplearning4j.spark.impl.common.Adder.(Adder.java:17) at org.deeplearning4j.spark.impl.multilayer.SparkDl4jMultiLayer.runIteration(SparkDl4jMultiLayer.java:237) at org.deeplearning4j.spark.impl.multilayer.SparkDl4jMultiLayer.fitDataSet(SparkDl4jMultiLayer.java:189) at poc.analaticsEngine.AnalaticsStockWorkBench$$anonfun$tra – Charls Joseph Nov 15 '16 at 08:43
  • I'm using spark 2.0.0 core jars. – Charls Joseph Nov 15 '16 at 08:46
  • Dl4j doesn't support spark 2.0 yet. – Adam Gibson Nov 19 '16 at 02:21