[GitHub] mariussoutier opened a new issue #12487: [Scala] PrefetchingIter fails with labels and predictions should have the same length

[GitBox]

mariussoutier opened a new issue #12487: [Scala] PrefetchingIter fails with 
labels and predictions should have the same length
URL: https://github.com/apache/incubator-mxnet/issues/12487
 
 
   ## Description
   
   Trying to use a `PrefetchingIter` to combine two `ImageRecordIter`. The two 
do not have the same number of images, but that shouldn't be a problem, right?
   
   ## Environment info (Required)
   
   macOS 10.13.6
   IntelliJ 2018.2.2
   Scala 2.11.12
   Java 1.8.0_121
   MXNet 1.2.1
   
   ## Error Message:
   
   ```scala
   Exception in thread "main" java.lang.IllegalArgumentException: requirement 
failed: labels and predictions should have the same length.
        at scala.Predef$.require(Predef.scala:224)
        at org.apache.mxnet.Accuracy.update(EvalMetric.scala:106)
        at 
org.apache.mxnet.DataParallelExecutorGroup$$anonfun$updateMetric$1.apply(ExecutorManager.scala:528)
        at 
org.apache.mxnet.DataParallelExecutorGroup$$anonfun$updateMetric$1.apply(ExecutorManager.scala:526)
        at 
scala.collection.IndexedSeqOptimized$class.foreach(IndexedSeqOptimized.scala:33)
        at scala.collection.mutable.ArrayOps$ofRef.foreach(ArrayOps.scala:186)
        at 
org.apache.mxnet.DataParallelExecutorGroup.updateMetric(ExecutorManager.scala:526)
        at 
org.apache.mxnet.DataParallelExecutorManager.updateMetric(ExecutorManager.scala:168)
        at 
org.apache.mxnet.Model$$anonfun$trainMultiDevice$1.apply$mcVI$sp(Model.scala:313)
        at scala.collection.immutable.Range.foreach$mVc$sp(Range.scala:160)
        at org.apache.mxnet.Model$.trainMultiDevice(Model.scala:284)
        at org.apache.mxnet.FeedForward.fit(FeedForward.scala:377)
        at org.apache.mxnet.FeedForward.fit(FeedForward.scala:281)
        at org.apache.mxnet.FeedForward$Builder.build(FeedForward.scala:676)
        at 
com.mariussoutier.example.TrainChars74K$.trainFF(TrainChars74K.scala:86)
        at 
com.mariussoutier.example.TrainChars74K$.main(TrainChars74K.scala:177)
        at com.mariussoutier.example.TrainChars74K.main(TrainChars74K.scala)
   ```
   
   
   ## Minimum reproducible example
   
   ```scala
       val trainDataIterA = IO.ImageRecordIter(Map(
         "data_name" -> dataName,
         "path_imgrec" -> 
this.getClass.getResource("/data/mydataA.rec").getFile,
         "data_shape" -> "(3,128,128)",
         "batch_size" -> "50",
         "rand_crop" -> "True",
          "shuffle" -> "False"
       ))
   
       val trainDataIterB = IO.ImageRecordIter(Map(
         "data_name" -> dataName,
         "path_imgrec" -> 
this.getClass.getResource("/data/mydataB.rec").getFile,
         "data_shape" -> "(3,128,128)",
         "batch_size" -> "50",
         "rand_crop" -> "True",
         "shuffle" -> "False"
       ))
   
           val trainDataIter = new PrefetchingIter(
             IndexedSeq(trainDataIterA, trainDataIterB)
           )
   ```
   

----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services