diff options
author | Christoph Sawade <christoph@sawade.me> | 2014-09-15 17:39:31 -0700 |
---|---|---|
committer | Xiangrui Meng <meng@databricks.com> | 2014-09-15 17:39:31 -0700 |
commit | 3b93128139e8d303f1d7bfd04e9a99a11a5b6404 (patch) | |
tree | a1239e81d6fae05f8727439080e7ce4f0f936de5 /mllib/src/main | |
parent | 37d925280cdfdda8f6f7174c67a614056eea5d69 (diff) | |
download | spark-3b93128139e8d303f1d7bfd04e9a99a11a5b6404.tar.gz spark-3b93128139e8d303f1d7bfd04e9a99a11a5b6404.tar.bz2 spark-3b93128139e8d303f1d7bfd04e9a99a11a5b6404.zip |
[SPARK-3396][MLLIB] Use SquaredL2Updater in LogisticRegressionWithSGD
SimpleUpdater ignores the regularizer, which leads to an unregularized
LogReg. To enable the common L2 regularizer (and the corresponding
regularization parameter) for logistic regression the SquaredL2Updater
has to be used in SGD (see, e.g., [SVMWithSGD])
Author: Christoph Sawade <christoph@sawade.me>
Closes #2398 from BigCrunsh/fix-regparam-logreg and squashes the following commits:
0820c04 [Christoph Sawade] Use SquaredL2Updater in LogisticRegressionWithSGD
Diffstat (limited to 'mllib/src/main')
-rw-r--r-- | mllib/src/main/scala/org/apache/spark/mllib/classification/LogisticRegression.scala | 2 |
1 files changed, 1 insertions, 1 deletions
diff --git a/mllib/src/main/scala/org/apache/spark/mllib/classification/LogisticRegression.scala b/mllib/src/main/scala/org/apache/spark/mllib/classification/LogisticRegression.scala index 486bdbfa9c..84d3c7cebd 100644 --- a/mllib/src/main/scala/org/apache/spark/mllib/classification/LogisticRegression.scala +++ b/mllib/src/main/scala/org/apache/spark/mllib/classification/LogisticRegression.scala @@ -84,7 +84,7 @@ class LogisticRegressionWithSGD private ( extends GeneralizedLinearAlgorithm[LogisticRegressionModel] with Serializable { private val gradient = new LogisticGradient() - private val updater = new SimpleUpdater() + private val updater = new SquaredL2Updater() override val optimizer = new GradientDescent(gradient, updater) .setStepSize(stepSize) .setNumIterations(numIterations) |