Github user jkbradley commented on a diff in the pull request:
https://github.com/apache/spark/pull/6504#discussion_r31369208
--- Diff: docs/ml-guide.md ---
@@ -157,6 +174,49 @@ There are now several algorithms in the Pipelines API
which are not in the lower
* [Feature Extraction, Transformation, and Selection](ml-features.html)
* [Ensembles](ml-ensembles.html)
+## Linear Methods with Elastic Net Regularization
+
+In MLlib, we implement popular linear methods such as logistic regression
and linear least squares with L1 or L2 regularization. Refer to [the linear
methods section](mllib-linear-methods.html) for details. In `spark.ml`, we add
the [Elastic net](http://users.stat.umn.edu/~zouxx019/Papers/elasticnet.pdf),
which is a hybrid of L1 and L2 regularization. Mathematically it is defined as
a linear combination of the L1-norm and the L2-norm:
--- End diff --
> In `spark.ml`, we add the [Elastic
net](http://users.stat.umn.edu/~zouxx019/Papers/elasticnet.pdf), which is a
hybrid of L1 and L2 regularization.
This makes it sound like it's only in spark.ml. Can this please instead
say that we provide a Pipelines API? The main thing needed here is a code
example, which should demonstrate how to do L1, L2, and a mix. (But I like the
note about how it uses a different optimizer.)
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastruct...@apache.org or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org
