[Weka] IndexOutOfBoundsException when obtaining folds

Start at org.openml.weka.experiment.TaskResultProducer.doRun, especially line 248, and trace how folds are obtained for Task 17 with Weka's NaiveBayes classifier. Reproduce the task using the reported dataset and inspect the surrounding TaskBasedExperiment flow; done means the IndexOutOfBoundsException no longer occurs when obtaining folds.

From @DraXus on February 14, 2015 15:17

The following error is shown when trying to run task 17 in Weka 3.7.12 using Naive Bayes as classifier.

[14-02-2015 13:41:27] [INFO] [ARFF Cache] Stored dataset 17 to cache.
[14-02-2015 13:41:27] [INFO] [ARFF Cache] Stored splits 17 to cache.
[14-02-2015 13:41:27] [INFO] [Splits] Obtaining folds for Task 17 (bridges) with weka.classifiers.bayes.NaiveBayes - Repeat 0
java.lang.IndexOutOfBoundsException: Index: 107, Size: 107
java.util.ArrayList.rangeCheck(ArrayList.java:635)
java.util.ArrayList.get(ArrayList.java:411)
weka.core.Instances.instance(Instances.java:768)
org.openml.weka.experiment.TaskResultProducer.doRun(TaskResultProducer.java:248)
org.openml.weka.experiment.TaskBasedExperiment.nextIteration(TaskBasedExperiment.java:173)
org.openml.weka.gui.OpenmlRunPanel$ExperimentRunner.run(OpenmlRunPanel.java:197)

at java.util.ArrayList.rangeCheck(ArrayList.java:635)
at java.util.ArrayList.get(ArrayList.java:411)
at weka.core.Instances.instance(Instances.java:768)
at org.openml.weka.experiment.TaskResultProducer.doRun(TaskResultProducer.java:248)
at org.openml.weka.experiment.TaskBasedExperiment.nextIteration(TaskBasedExperiment.java:173)
at org.openml.weka.gui.OpenmlRunPanel$ExperimentRunner.run(OpenmlRunPanel.java:197)

Copied from original issue: openml/OpenML#172