change gpu side

adRise · May 15, 2024 · c3ed781 · c3ed781
1 parent 1b20b25
commit c3ed781
Show file tree

Hide file tree

Showing 2 changed files with 24 additions and 1 deletion.
diff --git a/...rk-gpu/src/test/scala/ml/dmlc/xgboost4j/scala/rapids/spark/GpuXGBoostRegressorSuite.scala b/...rk-gpu/src/test/scala/ml/dmlc/xgboost4j/scala/rapids/spark/GpuXGBoostRegressorSuite.scala
@@ -23,8 +23,10 @@ import ml.dmlc.xgboost4j.scala.spark.{XGBoostRegressionModel, XGBoostRegressor}
 import org.apache.spark.ml.feature.VectorAssembler
 import org.apache.spark.sql.functions.{col, udf}
 import org.apache.spark.sql.types.{FloatType, IntegerType, StructField, StructType}
+import org.apache.commons.logging.LogFactory
 
 class GpuXGBoostRegressorSuite extends GpuTestSuite {
+  val logger = LogFactory.getLog("XGBoostSpark")
 
   val labelName = "label_col"
   val groupName = "group_col"
@@ -255,4 +257,22 @@ class GpuXGBoostRegressorSuite extends GpuTestSuite {
       assert(testDf.count() === ret.length)
     }
   }
+
+  test("Ranking: test position bias") {
+    withGpuSparkSession(enableCsvConf()) { spark =>
+      logger.info("hongfeili-scala: begin execute test")
+      val xgbParam = Map("eta" -> 0.1f, "max_depth" -> 2, "objective" -> "rank:ndcg",
+        "num_round" -> 10, "num_workers" -> 1, "tree_method" -> "gpu_hist",
+        "features_cols" -> featureNames, "label_col" -> labelName, "lambdarank_unbiased" -> true, "eval_metric" -> "ndcg")
+      val Array(trainingDf, testDf) = spark.read.option("header", "true").schema(schema)
+        .csv(getResourcePath("/rank.train.csv")).randomSplit(Array(0.7, 0.3), seed = 1)
+
+      val model = new XGBoostRegressor(xgbParam)
+        .setGroupCol(groupName)
+        .fit(trainingDf)
+
+      val ret = model.transform(testDf).collect()
+      assert(testDf.count() === ret.length)
+    }
+  }
 }
diff --git a/.../xgboost4j-spark/src/test/scala/ml/dmlc/xgboost4j/scala/spark/XGBoostRegressorSuite.scala b/.../xgboost4j-spark/src/test/scala/ml/dmlc/xgboost4j/scala/spark/XGBoostRegressorSuite.scala
@@ -26,9 +26,12 @@ import org.apache.spark.sql.{DataFrame, Row}
 import org.scalatest.funsuite.AnyFunSuite
 
 import org.apache.spark.ml.feature.VectorAssembler
+import org.apache.commons.logging.LogFactory
+
 
 class XGBoostRegressorSuite extends AnyFunSuite with PerTest with TmpFolderPerSuite {
   protected val treeMethod: String = "auto"
+  val logger = LogFactory.getLog("XGBoostSpark")
 
   test("XGBoost-Spark XGBoostRegressor output should match XGBoost4j") {
     val trainingDM = new DMatrix(Regression.train.iterator)
@@ -133,6 +136,7 @@ class XGBoostRegressorSuite extends AnyFunSuite with PerTest with TmpFolderPerSu
   }
 
   test("ranking: test position bias") {
+    logger.info("hongfeili-scala: begin execute test")
     val paramMap = Map("eta" -> "1", "max_depth" -> "6", "silent" -> "0", "verbosity" -> "3",
       "objective" -> "rank:ndcg", "num_workers" -> numWorkers, "num_round" -> 5,
       "group_col" -> "group", "tree_method" -> treeMethod, "lambdarank_unbiased" -> true, "eval_metric" -> "ndcg")
@@ -142,7 +146,6 @@ class XGBoostRegressorSuite extends AnyFunSuite with PerTest with TmpFolderPerSu
     val model = new XGBoostRegressor(paramMap).fit(trainingDF)
 
     val prediction = model.transform(testDF).collect()
-    println("hello---------hongfei")
     assert(testDF.count() === prediction.length)
   }