[SPARK-6457] Error when calling Pyspark RandomForestModel.load

• Type: Bug
• Status: Resolved
• Priority: Minor
• Resolution: Fixed
• Affects Version/s: 1.3.0
• Fix Version/s: 1.3.1, 1.4.0
• Component/s: MLlib, PySpark
• Assignee: Joseph K. Bradley
• Reporter: Joseph K. Bradley
• Created: 22/Mar/15 17:26
• Updated: 24/Apr/15 00:33
• Resolved: 23/Mar/15 02:31

Description

Reported by https://github.com/catmonkeylee:

Summary: PySpark RandomForestModel.load fails in test script. It appears that the saved model file is empty.

When I run the sample code in cluster mode, there is an error.

Traceback (most recent call last):
File "/data1/s/apps/spark-app/app/sample_rf.py", line 25, in
sameModel = RandomForestModel.load(sc, model_path)
File "/home/s/apps/spark/python/pyspark/mllib/util.py", line 254, in load
java_model = cls.load_java(sc, path)
File "/home/s/apps/spark/python/pyspark/mllib/util.py", line 250, in _load_java
return java_obj.load(sc._jsc.sc(), path)
File "/home/s/apps/spark/python/lib/py4j-0.8.2.1-src.zip/py4j/java_gateway.py", line 538, in __call
File "/home/s/apps/spark/python/lib/py4j-0.8.2.1-src.zip/py4j/protocol.py", line 300, in get_return_value
py4j.protocol.Py4JJavaError: An error occurred while calling z:org.apache.spark.mllib.tree.model.RandomForestModel.load.
: java.lang.UnsupportedOperationException: empty collection
at org.apache.spark.rdd.RDD.first(RDD.scala:1191)
at org.apache.spark.mllib.util.Loader$.loadMetadata(modelSaveLoad.scala:125)
at org.apache.spark.mllib.tree.model.RandomForestModel$.load(treeEnsembleModels.scala:65)
at org.apache.spark.mllib.tree.model.RandomForestModel.load(treeEnsembleModels.scala)
at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)

I run the code on a spark cluster , spark version is 1.3.0

The test code:
===================================
from pyspark import SparkContext, SparkConf
from pyspark.mllib.tree import RandomForest, RandomForestModel
from pyspark.mllib.util import MLUtils

conf = SparkConf().setAppName('LocalTest')
sc = SparkContext(conf=conf)
data = MLUtils.loadLibSVMFile(sc, 'data/mllib/sample_libsvm_data.txt')
print data.count()
(trainingData, testData) = data.randomSplit([0.7, 0.3])
model = RandomForest.trainClassifier(trainingData, numClasses=2, categoricalFeaturesInfo={},
numTrees=3, featureSubsetStrategy="auto",
impurity='gini', maxDepth=4, maxBins=32)

1. Evaluate model on test instances and compute test error
   predictions = model.predict(testData.map(lambda x: x.features))
   labelsAndPredictions = testData.map(lambda lp: lp.label).zip(predictions)
   testErr = labelsAndPredictions.filter(lambda (v, p): v != p).count() / float(testData.count())
   print('Test Error = ' + str(testErr))
   print('Learned classification forest model:')
   print(model.toDebugString())

1. Save and load model
   _model_path = "/home/s/apps/spark-app/data/myModelPath"
   model.save(sc, _model_path)
   sameModel = RandomForestModel.load(sc, _model_path)
   sc.stop()

===================
run command:
spark-submit --master spark://t0.q.net:7077 --executor-memory 1G sample_rf.py

======================
Then I get this error :

Traceback (most recent call last):
File "/data1/s/apps/spark-app/app/sample_rf.py", line 25, in <module>
sameModel = RandomForestModel.load(sc, _model_path)
File "/home/s/apps/spark/python/pyspark/mllib/util.py", line 254, in load
java_model = cls._load_java(sc, path)
File "/home/s/apps/spark/python/pyspark/mllib/util.py", line 250, in _load_java
return java_obj.load(sc._jsc.sc(), path)
File "/home/s/apps/spark/python/lib/py4j-0.8.2.1-src.zip/py4j/java_gateway.py", line 538, in _call_
File "/home/s/apps/spark/python/lib/py4j-0.8.2.1-src.zip/py4j/protocol.py", line 300, in get_return_value
py4j.protocol.Py4JJavaError: An error occurred while calling z:org.apache.spark.mllib.tree.model.RandomForestModel.load.
: java.lang.UnsupportedOperationException: empty collection
at org.apache.spark.rdd.RDD.first(RDD.scala:1191)
at org.apache.spark.mllib.util.Loader$.loadMetadata(modelSaveLoad.scala:125)
at org.apache.spark.mllib.tree.model.RandomForestModel$.load(treeEnsembleModels.scala:65)
at org.apache.spark.mllib.tree.model.RandomForestModel.load(treeEnsembleModels.scala)
at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:57)
at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43)
at java.lang.reflect.Method.invoke(Method.java:606)
at py4j.reflection.MethodInvoker.invoke(MethodInvoker.java:231)
at py4j.reflection.ReflectionEngine.invoke(ReflectionEngine.java:379)
at py4j.Gateway.invoke(Gateway.java:259)
at py4j.commands.AbstractCommand.invokeMethod(AbstractCommand.java:133)
at py4j.commands.CallCommand.execute(CallCommand.java:79)
at py4j.GatewayConnection.run(GatewayConnection.java:207)
at java.lang.Thread.run(Thread.java:724)

Issue Links

• requires: SPARK-6330 newParquetRelation gets incorrect FileSystem