perf: improve lightgbm training performance 4x-10x by setting num_thr…

…eads to be cores-1 (#1282)
microsoft · Dec 3, 2021 · ff2aa02 · ff2aa02
1 parent 6ea8a9a
commit ff2aa02
Show file tree

Hide file tree

Showing 4 changed files with 9 additions and 5 deletions.
diff --git a/lightgbm/src/main/scala/com/microsoft/azure/synapse/ml/lightgbm/LightGBMBase.scala b/lightgbm/src/main/scala/com/microsoft/azure/synapse/ml/lightgbm/LightGBMBase.scala
@@ -247,8 +247,12 @@ trait LightGBMBase[TrainedModel <: Model[TrainedModel]] extends Estimator[Traine
     *
     * @return ExecutionParams object containing parameters related to LightGBM execution.
     */
-  protected def getExecutionParams: ExecutionParams = {
-    ExecutionParams(getChunkSize, getMatrixType, getNumThreads, getUseSingleDatasetMode)
+  protected def getExecutionParams(numTasksPerExec: Int): ExecutionParams = {
+    val execNumThreads =
+      if (getUseSingleDatasetMode) get(numThreads).getOrElse(numTasksPerExec - 1)
+      else getNumThreads
+
+    ExecutionParams(getChunkSize, getMatrixType, execNumThreads, getUseSingleDatasetMode)
   }
 
   protected def getColumnParams: ColumnParams = {

diff --git a/lightgbm/src/main/scala/com/microsoft/azure/synapse/ml/lightgbm/LightGBMClassifier.scala b/lightgbm/src/main/scala/com/microsoft/azure/synapse/ml/lightgbm/LightGBMClassifier.scala
@@ -54,7 +54,7 @@ class LightGBMClassifier(override val uid: String)
       getIsUnbalance, getVerbosity, categoricalIndexes, actualNumClasses, getBoostFromAverage,
       getBoostingType, get(lambdaL1), get(lambdaL2), get(isProvideTrainingMetric),
       get(metric), get(minGainToSplit), get(maxDeltaStep), getMaxBinByFeature, get(minDataInLeaf), getSlotNames,
-      getDelegate, getDartParams, getExecutionParams, getObjectiveParams)
+      getDelegate, getDartParams, getExecutionParams(numTasksPerExec), getObjectiveParams)
   }
 
   def getModel(trainParams: TrainParams, lightGBMBooster: LightGBMBooster): LightGBMClassificationModel = {

diff --git a/lightgbm/src/main/scala/com/microsoft/azure/synapse/ml/lightgbm/LightGBMRanker.scala b/lightgbm/src/main/scala/com/microsoft/azure/synapse/ml/lightgbm/LightGBMRanker.scala
@@ -60,7 +60,7 @@ class LightGBMRanker(override val uid: String)
       getVerbosity, categoricalIndexes, getBoostingType, get(lambdaL1), get(lambdaL2), getMaxPosition, getLabelGain,
       get(isProvideTrainingMetric), get(metric), getEvalAt, get(minGainToSplit), get(maxDeltaStep),
       getMaxBinByFeature, get(minDataInLeaf), getSlotNames, getDelegate, getDartParams,
-      getExecutionParams, getObjectiveParams)
+      getExecutionParams(numTasksPerExec), getObjectiveParams)
   }
 
   def getModel(trainParams: TrainParams, lightGBMBooster: LightGBMBooster): LightGBMRankerModel = {

diff --git a/lightgbm/src/main/scala/com/microsoft/azure/synapse/ml/lightgbm/LightGBMRegressor.scala b/lightgbm/src/main/scala/com/microsoft/azure/synapse/ml/lightgbm/LightGBMRegressor.scala
@@ -70,7 +70,7 @@ class LightGBMRegressor(override val uid: String)
       getBoostFromAverage, getBoostingType, get(lambdaL1), get(lambdaL2), get(isProvideTrainingMetric),
       get(metric), get(minGainToSplit), get(maxDeltaStep),
       getMaxBinByFeature, get(minDataInLeaf), getSlotNames, getDelegate,
-      getDartParams, getExecutionParams, getObjectiveParams)
+      getDartParams, getExecutionParams(numTasksPerExec), getObjectiveParams)
   }
 
   def getModel(trainParams: TrainParams, lightGBMBooster: LightGBMBooster): LightGBMRegressionModel = {