[SPARK-3412] [PySpark] Replace Epydoc with Sphinx to generate Python API docs

Retire Epydoc, use Sphinx to generate API docs.

Refine Sphinx docs, also convert some docstrings into Sphinx style.

It looks like:
![api doc](https://cloud.githubusercontent.com/assets/40902/4538272/9e2d4f10-4dec-11e4-8d96-6e45a8fe51f9.png)

Author: Davies Liu <davies.liu@gmail.com>

Closes #2689 from davies/docs and squashes the following commits:

bf4a0a5 [Davies Liu] fix links
3fb1572 [Davies Liu] fix _static in jekyll
65a287e [Davies Liu] fix scripts and logo
8524042 [Davies Liu] Merge branch 'master' of github.com:apache/spark into docs
d5b874a [Davies Liu] Merge branch 'master' of github.com:apache/spark into docs
4bc1c3c [Davies Liu] refactor
746d0b6 [Davies Liu] @param -> :param
240b393 [Davies Liu] replace epydoc with sphinx doc

1 files changed, 16 insertions, 16 deletions

diff --git a/python/pyspark/mllib/classification.py b/python/pyspark/mllib/classification.py
index a765b1c4f7..cd43982191 100644
--- a/python/pyspark/mllib/classification.py
+++ b/python/pyspark/mllib/classification.py
@@ -79,15 +79,15 @@ class LogisticRegressionWithSGD(object):
         """
         Train a logistic regression model on the given data.
 
-        @param data:              The training data.
-        @param iterations:        The number of iterations (default: 100).
-        @param step:              The step parameter used in SGD
+        :param data:              The training data.
+        :param iterations:        The number of iterations (default: 100).
+        :param step:              The step parameter used in SGD
                                   (default: 1.0).
-        @param miniBatchFraction: Fraction of data to be used for each SGD
+        :param miniBatchFraction: Fraction of data to be used for each SGD
                                   iteration.
-        @param initialWeights:    The initial weights (default: None).
-        @param regParam:          The regularizer parameter (default: 1.0).
-        @param regType:           The type of regularizer used for training
+        :param initialWeights:    The initial weights (default: None).
+        :param regParam:          The regularizer parameter (default: 1.0).
+        :param regType:           The type of regularizer used for training
                                   our model.
 
                                   :Allowed values:
@@ -151,15 +151,15 @@ class SVMWithSGD(object):
         """
         Train a support vector machine on the given data.
 
-        @param data:              The training data.
-        @param iterations:        The number of iterations (default: 100).
-        @param step:              The step parameter used in SGD
+        :param data:              The training data.
+        :param iterations:        The number of iterations (default: 100).
+        :param step:              The step parameter used in SGD
                                   (default: 1.0).
-        @param regParam:          The regularizer parameter (default: 1.0).
-        @param miniBatchFraction: Fraction of data to be used for each SGD
+        :param regParam:          The regularizer parameter (default: 1.0).
+        :param miniBatchFraction: Fraction of data to be used for each SGD
                                   iteration.
-        @param initialWeights:    The initial weights (default: None).
-        @param regType:           The type of regularizer used for training
+        :param initialWeights:    The initial weights (default: None).
+        :param regType:           The type of regularizer used for training
                                   our model.
 
                                   :Allowed values:
@@ -238,10 +238,10 @@ class NaiveBayes(object):
         classification.  By making every vector a 0-1 vector, it can also be
         used as Bernoulli NB (U{http://tinyurl.com/p7c96j6}).
 
-        @param data: RDD of NumPy vectors, one per element, where the first
+        :param data: RDD of NumPy vectors, one per element, where the first
                coordinate is the label and the rest is the feature vector
                (e.g. a count vector).
-        @param lambda_: The smoothing parameter
+        :param lambda_: The smoothing parameter
         """
         sc = data.context
         jlist = sc._jvm.PythonMLLibAPI().trainNaiveBayes(data._to_java_object_rdd(), lambda_)