[SPARK-14829][MLLIB] Deprecate GLM APIs using SGD

## What changes were proposed in this pull request?
According to the [SPARK-14829](https://issues.apache.org/jira/browse/SPARK-14829), deprecate API of LogisticRegression and LinearRegression using SGD

## How was this patch tested?
manual tests

Author: Zheng RuiFeng <ruifengz@foxmail.com>

Closes #12596 from zhengruifeng/deprecate_sgd.

1 files changed, 18 insertions, 0 deletions

diff --git a/python/pyspark/mllib/regression.py b/python/pyspark/mllib/regression.py
index 3b77a62000..639c5eabaa 100644
--- a/python/pyspark/mllib/regression.py
+++ b/python/pyspark/mllib/regression.py
@@ -17,6 +17,7 @@
 
 import numpy as np
 from numpy import array
+import warnings
 
 from pyspark import RDD, since
 from pyspark.streaming.dstream import DStream
@@ -221,6 +222,7 @@ def _regression_train_wrapper(train_func, modelClass, data, initial_weights):
 class LinearRegressionWithSGD(object):
     """
     .. versionadded:: 0.9.0
+    .. note:: Deprecated in 2.0.0. Use ml.regression.LinearRegression.
     """
     @classmethod
     @since("0.9.0")
@@ -276,6 +278,8 @@ class LinearRegressionWithSGD(object):
           A condition which decides iteration termination.
           (default: 0.001)
         """
+        warnings.warn("Deprecated in 2.0.0. Use ml.regression.LinearRegression.")
+
         def train(rdd, i):
             return callMLlibFunc("trainLinearRegressionModelWithSGD", rdd, int(iterations),
                                  float(step), float(miniBatchFraction), i, float(regParam),
@@ -366,6 +370,8 @@ class LassoModel(LinearRegressionModelBase):
 class LassoWithSGD(object):
     """
     .. versionadded:: 0.9.0
+    .. note:: Deprecated in 2.0.0. Use ml.regression.LinearRegression with elasticNetParam = 1.0.
+            Note the default regParam is 0.01 for LassoWithSGD, but is 0.0 for LinearRegression.
     """
     @classmethod
     @since("0.9.0")
@@ -413,6 +419,10 @@ class LassoWithSGD(object):
           A condition which decides iteration termination.
           (default: 0.001)
         """
+        warnings.warn(
+            "Deprecated in 2.0.0. Use ml.regression.LinearRegression with elasticNetParam = 1.0. "
+            "Note the default regParam is 0.01 for LassoWithSGD, but is 0.0 for LinearRegression.")
+
         def train(rdd, i):
             return callMLlibFunc("trainLassoModelWithSGD", rdd, int(iterations), float(step),
                                  float(regParam), float(miniBatchFraction), i, bool(intercept),
@@ -503,6 +513,9 @@ class RidgeRegressionModel(LinearRegressionModelBase):
 class RidgeRegressionWithSGD(object):
     """
     .. versionadded:: 0.9.0
+    .. note:: Deprecated in 2.0.0. Use ml.regression.LinearRegression with elasticNetParam = 0.0.
+            Note the default regParam is 0.01 for RidgeRegressionWithSGD, but is 0.0 for
+            LinearRegression.
     """
     @classmethod
     @since("0.9.0")
@@ -550,6 +563,11 @@ class RidgeRegressionWithSGD(object):
           A condition which decides iteration termination.
           (default: 0.001)
         """
+        warnings.warn(
+            "Deprecated in 2.0.0. Use ml.regression.LinearRegression with elasticNetParam = 0.0. "
+            "Note the default regParam is 0.01 for RidgeRegressionWithSGD, but is 0.0 for "
+            "LinearRegression.")
+
         def train(rdd, i):
             return callMLlibFunc("trainRidgeModelWithSGD", rdd, int(iterations), float(step),
                                  float(regParam), float(miniBatchFraction), i, bool(intercept),