Github user dbtsai commented on a diff in the pull request:
https://github.com/apache/spark/pull/10940#discussion_r51051112
--- Diff:
mllib/src/main/scala/org/apache/spark/ml/classification/LogisticRegression.scala
---
@@ -341,11 +341,11 @@ class LogisticRegression @Since("1.2.0") (
regParamL1
} else {
// If `standardization` is false, we still standardize the
data
- // to improve the rate of convergence; as a result, we
have to
- // perform this reverse standardization by penalizing each
component
- // differently to get effectively the same objective
function when
+ // to improve the rate of convergence unless the standard
deviation is zero;
+ // as a result, we have to perform this reverse
standardization by penalizing
+ // each component differently to get effectively the same
objective function when
// the training dataset is not standardized.
- if (featuresStd(index) != 0.0) regParamL1 /
featuresStd(index) else 0.0
+ if (featuresStd(index) != 0.0) regParamL1 /
featuresStd(index) else regParamL1
--- End diff --
The constant `value` can be really large or very small negatively. The
optimizer may not be able to converge well in this case. I don't prove or try
it yet, but mathematically, with the following changes, this should be solving
identical problem.
```scala
// the training dataset is not standardized.
if (featuresStd(index) != 0.0) regParamL1 / featuresStd(index) else
regParamL1 / featuresMean(index)
```
---
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
