Skip to content
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.

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

[jvm-packages] fix the prediction issue for multi:softmax #7694

Merged
merged 1 commit into from
Feb 23, 2022
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
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
Original file line number Diff line number Diff line change
Expand Up @@ -385,35 +385,49 @@ class XGBoostClassificationModel private[ml](
Vectors.dense(rawPredictions)
}

val probabilityUDF = udf { probability: mutable.WrappedArray[Float] =>
val prob = probability.map(_.toDouble).toArray
val probabilities = if (numClasses == 2) Array(1.0 - prob(0), prob(0)) else prob
Vectors.dense(probabilities)
}

val predictUDF = udf { probability: mutable.WrappedArray[Float] =>
// From XGBoost probability to MLlib prediction
val prob = probability.map(_.toDouble).toArray
val probabilities = if (numClasses == 2) Array(1.0 - prob(0), prob(0)) else prob
probability2prediction(Vectors.dense(probabilities))
}

if ($(rawPredictionCol).nonEmpty) {
outputData = outputData
.withColumn(getRawPredictionCol, rawPredictionUDF(col(_rawPredictionCol)))
numColsOutput += 1
}

if ($(probabilityCol).nonEmpty) {
outputData = outputData
.withColumn(getProbabilityCol, probabilityUDF(col(_probabilityCol)))
numColsOutput += 1
}
if (getObjective.equals("multi:softmax")) {
// For objective=multi:softmax scenario, there is no probability predicted from xgboost.
// Instead, the probability column will be filled with real prediction
val predictUDF = udf { probability: mutable.WrappedArray[Float] =>
probability(0)
}
if ($(predictionCol).nonEmpty) {
outputData = outputData
.withColumn($(predictionCol), predictUDF(col(_probabilityCol)))
numColsOutput += 1
}

if ($(predictionCol).nonEmpty) {
outputData = outputData
.withColumn($(predictionCol), predictUDF(col(_probabilityCol)))
numColsOutput += 1
} else {
val probabilityUDF = udf { probability: mutable.WrappedArray[Float] =>
val prob = probability.map(_.toDouble).toArray
val probabilities = if (numClasses == 2) Array(1.0 - prob(0), prob(0)) else prob
Vectors.dense(probabilities)
}
if ($(probabilityCol).nonEmpty) {
outputData = outputData
.withColumn(getProbabilityCol, probabilityUDF(col(_probabilityCol)))
numColsOutput += 1
}

val predictUDF = udf { probability: mutable.WrappedArray[Float] =>
// From XGBoost probability to MLlib prediction
val prob = probability.map(_.toDouble).toArray
val probabilities = if (numClasses == 2) Array(1.0 - prob(0), prob(0)) else prob
probability2prediction(Vectors.dense(probabilities))
}
if ($(predictionCol).nonEmpty) {
outputData = outputData
.withColumn($(predictionCol), predictUDF(col(_probabilityCol)))
numColsOutput += 1
}
}

if (numColsOutput == 0) {
Expand Down
Original file line number Diff line number Diff line change
@@ -1,5 +1,5 @@
/*
Copyright (c) 2014 by Contributors
Copyright (c) 2014-2022 by Contributors

Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
Expand All @@ -17,9 +17,11 @@
package ml.dmlc.xgboost4j.scala.spark

import ml.dmlc.xgboost4j.scala.{DMatrix, XGBoost => ScalaXGBoost}

import org.apache.spark.ml.linalg._
import org.apache.spark.sql._
import org.scalatest.FunSuite

import org.apache.spark.Partitioner

class XGBoostClassifierSuite extends FunSuite with PerTest {
Expand Down Expand Up @@ -102,6 +104,8 @@ class XGBoostClassifierSuite extends FunSuite with PerTest {
assert(model.getEta == 0.1)
assert(model.getMaxDepth == 6)
assert(model.numClasses == 6)
val transformedDf = model.transform(trainingDF)
assert(!transformedDf.columns.contains("probability"))
}

test("use base margin") {
Expand Down