Skip to content

[SPARK-32455][ML][Follow-Up] LogisticRegressionModel prediction optimization - fix incorrect initialization #30013

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Closed
zhengruifeng wants to merge 3 commits into from
Closed

[SPARK-32455][ML][Follow-Up] LogisticRegressionModel prediction optimization - fix incorrect initialization #30013

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
27eab00
check code
zhengruifeng Oct 12, 2020
49690eb
use lazy array
zhengruifeng Oct 12, 2020
7a1749c
update testsuite
zhengruifeng Oct 13, 2020
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
26 changes: 15 additions & 11 deletions mllib/src/main/scala/org/apache/spark/ml/classification/LogisticRegression.scala
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1100,28 +1100,32 @@ class LogisticRegressionModel private[spark] (

private lazy val _intercept = interceptVector(0)
private lazy val _interceptVector = interceptVector.toDense
private var _threshold = Double.NaN
private var _rawThreshold = Double.NaN

updateBinaryThreshold()
private lazy val _binaryThresholdArray = {
val array = Array(Double.NaN, Double.NaN)
updateBinaryThresholds(array)
array
}
private def _threshold: Double = _binaryThresholdArray(0)
private def _rawThreshold: Double = _binaryThresholdArray(1)

private def updateBinaryThreshold(): Unit = {
private def updateBinaryThresholds(array: Array[Double]): Unit = {
if (!isMultinomial) {
_threshold = getThreshold
val _threshold = getThreshold
array(0) = _threshold
if (_threshold == 0.0) {
_rawThreshold = Double.NegativeInfinity
array(1) = Double.NegativeInfinity
} else if (_threshold == 1.0) {
_rawThreshold = Double.PositiveInfinity
array(1) = Double.PositiveInfinity
} else {
_rawThreshold = math.log(_threshold / (1.0 - _threshold))
array(1) = math.log(_threshold / (1.0 - _threshold))
}
}
}

@Since("1.5.0")
override def setThreshold(value: Double): this.type = {
super.setThreshold(value)
updateBinaryThreshold()
updateBinaryThresholds(_binaryThresholdArray)
this
}

Expand All @@ -1131,7 +1135,7 @@ class LogisticRegressionModel private[spark] (
@Since("1.5.0")
override def setThresholds(value: Array[Double]): this.type = {
super.setThresholds(value)
updateBinaryThreshold()
updateBinaryThresholds(_binaryThresholdArray)
this
}

Expand Down
3 changes: 1 addition & 2 deletions mllib/src/test/scala/org/apache/spark/ml/classification/LogisticRegressionSuite.scala
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -400,10 +400,9 @@ class LogisticRegressionSuite extends MLTest with DefaultReadWriteTest {
}

test("thresholds prediction") {
val blr = new LogisticRegression().setFamily("binomial")
val blr = new LogisticRegression().setFamily("binomial").setThreshold(1.0)
val binaryModel = blr.fit(smallBinaryDataset)

binaryModel.setThreshold(1.0)
testTransformer[(Double, Vector)](smallBinaryDataset.toDF(), binaryModel, "prediction") {
Copy link
Contributor Author

@zhengruifeng zhengruifeng Oct 13, 2020

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

master fails in this modified case:
tx

row => assert(row.getDouble(0) === 0.0)
}
Expand Down