Export - CSV (All fields)
Export - CSV (Current fields)
SPARK-5148
Make usersOut/productsOut storagelevel in ALS configurable
Use SparkConf#getBoolean instead of get().toBooleanSPARK-5667
Remove version from spark-ec2 example.
RowMatrix easily gets int overflow in the memory size warning
Provide isEmpty() function in RDD API
examples for ml don't have sparkContext.stop
Remove unreachable driver memory properties in yarn client mode (YarnClientSchedulerBackend)
add sc.stop to LDA examples
PythonMLlibAPI trainGaussianMixture seed should use Java type
Use python doc "note" for experimental tags in tree.py
Incorrect DataFrame Documentation Type References
Rename "externalSorting" in Aggregator
SPARK-4733
Add missing prameter comments in ShuffleDependency
Vectors.sqdist return inconsistent result for sparse/dense vectors when the vectors have different lengths
Scripts do not use SPARK_CONF_DIR where they should
Remove Shark from spark-ec2
Time major blocks of code in spark-ec2/setup.sh
`create table` DDL need to check if table exists first
Local Matrix apply should be public
ScalaReflection.convertToCatalyst should support nested arraysSPARK-4494
IDFModel.transform() add support for single vector
Add exponential, log normal, and gamma distributions to data generator to PySpark's MLlib- SPARK-4889
HistoryServer documentation refers to non-existent spark-history-server.sh script 
SPARK-4855
Python tests for hypothesis testing
way to find assembly jar is too strict
BlockManager should check whether blocks have already been removed Checking block is null or not after having gotten info.lock in remove block method
The name of Parquet File generated by AppendingParquetOutputFormat should be zero padded
OOM when making bins in BinaryClassificationMetrics
When the vocabulary size is large, Word2Vec may yield "OutOfMemoryError: Requested array size exceeds VM limit"
Pass maxIterations in as a parameter in Analyzer
Create SchemaRDD from case classes with multiple constructors- SPARK-4822
Use sphinx tags for Python doc annotations
SVD should check for k < 1SPARK-18306
CLONE - Support model save/load in Python's ALS.
Add numRows param in DataFrame.show- SPARK-5958
Update block matrix user guide - SPARK-5974
Add save/load to examples in ML guide - SPARK-5805
Fix the type error in the final example given in MLlib - Clustering documentation
Add an option to print the spark version in spark script
YarnClientSchedulerBackend fails to get application report when yarn restarts
Graph should have isCheckpointed, getCheckpointFiles methods
LocalLAPACK mode in RowMatrix.computeSVD should have much smaller upper bound
Use spark-class for running executors directly on mesos- SPARK-6083
Make Python API example consistent in NaiveBayes
NoSuchMethodError in Spark app is swallowed by YARN AM- SPARK-5892
Clean up ML, MLlib docs for 1.3 release
Run the PIC algorithm with degree vector
Add support for WITH SERDEPROPERTIES, TBLPROPERTIES in CREATE TEMPORARY TABLE
Replace Vector.toBreeze.activeIterator with foreachActive
Support model save/load in Python's ALS.