[SPARK-4127] [MLlib] [PySpark] Python bindings for StreamingLinearRegressionWithSGD

[MechCoder]

Python bindings for StreamingLinearRegressionWithSGD

[MechCoder]

cc: @freeman-lab @davies Will remove the LinearDataGenerator stuff (or rebase over master) as soon as #6715 is merged.

[SparkQA]

Test build #34596 has finished for PR 6744 at commit fdaa8b0.

• This patch fails Python style tests.
• This patch merges cleanly.
• This patch adds the following public classes (experimental):
  • class StreamingLinearRegressionWithSGD(LinearRegressionModel):
  • class LinearDataGenerator(object):

[SparkQA]

Test build #34602 has finished for PR 6744 at commit 8a0e560.

• This patch fails PySpark unit tests.
• This patch merges cleanly.
• This patch adds the following public classes (experimental):
  • class StreamingLinearRegressionWithSGD(LinearRegressionModel):
  • class LinearDataGenerator(object):

[SparkQA]

Test build #34672 has finished for PR 6744 at commit e6c5299.

• This patch passes all tests.
• This patch merges cleanly.
• This patch adds the following public classes (experimental):
  • class StreamingLinearRegressionWithSGD(LinearRegressionModel):
  • class LinearDataGenerator(object):

[SparkQA]

Test build #34979 has finished for PR 6744 at commit 30bac18.

• This patch fails PySpark unit tests.
• This patch merges cleanly.
• This patch adds the following public classes (experimental):
  • class StreamingLinearRegressionWithSGD(LinearRegressionModel):
  • class LinearDataGenerator(object):

[SparkQA]

Test build #35004 has finished for PR 6744 at commit c9aaf7e.

• This patch passes all tests.
• This patch merges cleanly.
• This patch adds the following public classes (experimental):
  • class StreamingLinearRegressionWithSGD(LinearRegressionModel):
  • class LinearDataGenerator(object):

[SparkQA]

Test build #35692 has finished for PR 6744 at commit cdad794.

• This patch passes all tests.
• This patch merges cleanly.
• This patch adds the following public classes (experimental):
  • class StreamingLinearRegressionWithSGD(LinearRegressionModel):

[SparkQA]

Test build #35697 has finished for PR 6744 at commit 5675a16.

• This patch passes all tests.
• This patch merges cleanly.
• This patch adds the following public classes (experimental):
  • class StreamingLinearRegressionWithSGD(LinearRegressionModel):

[freeman-lab]

stream -> batch

[MechCoder]

@freeman-lab @mengxr I rebased over master and inherited from StreamingLinearAlgorithm to avoid code duplication. Please give another pass.

[SparkQA]

Test build #35780 has finished for PR 6744 at commit 618a93b.

• This patch fails PySpark unit tests.
• This patch merges cleanly.
• This patch adds no public classes.

[SparkQA]

Test build #35782 has finished for PR 6744 at commit d47cc24.

• This patch fails PySpark unit tests.
• This patch merges cleanly.
• This patch adds the following public classes (experimental):
  • class StreamingLinearRegressionWithSGD(StreamingLinearAlgorithm):

[SparkQA]

Test build #35779 has finished for PR 6744 at commit 1b4ddd6.

• This patch fails PySpark unit tests.
• This patch merges cleanly.
• This patch adds the following public classes (experimental):
  • class StreamingLinearRegressionWithSGD(LinearRegressionModel):

[SparkQA]

Test build #35784 has finished for PR 6744 at commit d8f6457.

• This patch passes all tests.
• This patch merges cleanly.
• This patch adds the following public classes (experimental):
  • class StreamingLinearAlgorithm(object):
  • class StreamingLinearRegressionWithSGD(StreamingLinearAlgorithm):

[MechCoder]

I moved StreamingLinearAlgorithm to regression.py so that it complies with LabeledPoint and (there were circular imports)

[MechCoder]

jenkins retest this please

[mengxr]

LGTM. Merged into master. Thanks! (The last Jenkins build was successful.)

[SparkQA]

Test build #36161 has finished for PR 6744 at commit d8f6457.

• This patch passes all tests.
• This patch merges cleanly.
• This patch adds the following public classes (experimental):
  • class StreamingLinearAlgorithm(object):
  • class StreamingLinearRegressionWithSGD(StreamingLinearAlgorithm):