diff options
author | Davies Liu <davies.liu@gmail.com> | 2014-10-16 14:56:50 -0700 |
---|---|---|
committer | Xiangrui Meng <meng@databricks.com> | 2014-10-16 14:56:50 -0700 |
commit | 091d32c52e9d73da95896016c1d920e89858abfa (patch) | |
tree | 904edd29e64b57fa1ab72d3ca37ed2996aa9d1e4 /python/pyspark/mllib/util.py | |
parent | 4c589cac4496c6a4bb8485a340bd0641dca13847 (diff) | |
download | spark-091d32c52e9d73da95896016c1d920e89858abfa.tar.gz spark-091d32c52e9d73da95896016c1d920e89858abfa.tar.bz2 spark-091d32c52e9d73da95896016c1d920e89858abfa.zip |
[SPARK-3971] [MLLib] [PySpark] hotfix: Customized pickler should work in cluster mode
Customized pickler should be registered before unpickling, but in executor, there is no way to register the picklers before run the tasks.
So, we need to register the picklers in the tasks itself, duplicate the javaToPython() and pythonToJava() in MLlib, call SerDe.initialize() before pickling or unpickling.
Author: Davies Liu <davies.liu@gmail.com>
Closes #2830 from davies/fix_pickle and squashes the following commits:
0c85fb9 [Davies Liu] revert the privacy change
6b94e15 [Davies Liu] use JavaConverters instead of JavaConversions
0f02050 [Davies Liu] hotfix: Customized pickler does not work in cluster
Diffstat (limited to 'python/pyspark/mllib/util.py')
-rw-r--r-- | python/pyspark/mllib/util.py | 6 |
1 files changed, 3 insertions, 3 deletions
diff --git a/python/pyspark/mllib/util.py b/python/pyspark/mllib/util.py index 1357fd4fbc..84b39a4861 100644 --- a/python/pyspark/mllib/util.py +++ b/python/pyspark/mllib/util.py @@ -19,7 +19,7 @@ import numpy as np import warnings from pyspark.rdd import RDD -from pyspark.serializers import BatchedSerializer, PickleSerializer +from pyspark.serializers import AutoBatchedSerializer, PickleSerializer from pyspark.mllib.linalg import Vectors, SparseVector, _convert_to_vector from pyspark.mllib.regression import LabeledPoint @@ -174,8 +174,8 @@ class MLUtils(object): """ minPartitions = minPartitions or min(sc.defaultParallelism, 2) jrdd = sc._jvm.PythonMLLibAPI().loadLabeledPoints(sc._jsc, path, minPartitions) - jpyrdd = sc._jvm.PythonRDD.javaToPython(jrdd) - return RDD(jpyrdd, sc, BatchedSerializer(PickleSerializer())) + jpyrdd = sc._jvm.SerDe.javaToPython(jrdd) + return RDD(jpyrdd, sc, AutoBatchedSerializer(PickleSerializer())) def _test(): |