Remove duplicate test data in xgboost-predictor-example

xgboost-predictor-examples/build.gradle

@@ -3,5 +3,6 @@ apply plugin: 'scala'
 dependencies {
     compile project(":xgboost-predictor")
     compile project(":xgboost-predictor-spark${scalaSuffix}")
+    compile project(":xgboost-predictor-test")
     compile group: 'org.apache.spark', name: "spark-mllib${scalaSuffix}", version: sparkVersion
 }

xgboost-predictor-examples/src/main/java/biz/k11i/xgboost/demo/Example.java

@@ -1,6 +1,7 @@
 package biz.k11i.xgboost.demo;
 
 import biz.k11i.xgboost.Predictor;
+import biz.k11i.xgboost.TestHelper;
 import biz.k11i.xgboost.util.FVec;
 
 import java.io.File;
@@ -17,7 +18,7 @@
 public class Example {
     public static void main(String[] args) throws IOException {
         List<SimpleEntry<Integer, FVec>> data = loadData();
-        Predictor predictor = new Predictor(Example.class.getResourceAsStream("model/binary-logistic.model"));
+        Predictor predictor = new Predictor(TestHelper.getResourceAsStream("model/gbtree/v47/binary-logistic.model"));
 
         predictAndLogLoss(predictor, data);
 
@@ -74,7 +75,7 @@ static void predictLeafIndex(Predictor predictor, List<SimpleEntry<Integer, FVec
     static List<SimpleEntry<Integer, FVec>> loadData() throws IOException {
         List<SimpleEntry<Integer, FVec>> result = new ArrayList<>();
 
-        for (String line : Files.readAllLines(new File(Example.class.getResource("model/agaricus.txt.test").getPath()).toPath(), StandardCharsets.UTF_8)) {
+        for (String line : Files.readAllLines(new File(TestHelper.getResourcePath("data/agaricus.txt.0.test")).toPath(), StandardCharsets.UTF_8)) {
             String[] values = line.split(" ");
 
             Map<Integer, Float> map = new HashMap<>();

xgboost-predictor-examples/src/main/scala/biz/k11i/xgboost/spark/demo/BinaryClassificationExample.scala

@@ -1,5 +1,6 @@
 package biz.k11i.xgboost.spark.demo
 
+import biz.k11i.xgboost.TemporaryFileResource
 import biz.k11i.xgboost.spark.model.XGBoostBinaryClassification
 import org.apache.spark.SparkConf
 import org.apache.spark.ml.evaluation.BinaryClassificationEvaluator
@@ -10,9 +11,18 @@ object BinaryClassificationExample {
     val sparkConf = new SparkConf().setAppName("BinaryClassificationExample")
       .setMaster("local")
     val sparkSession = SparkSession.builder().config(sparkConf).getOrCreate()
+    val tempFileResource = new TemporaryFileResource
 
-    val modelPath = getResourcePath("/biz/k11i/xgboost/demo/model/spark/agaricus.model.spark")
-    val testDataPath = getResourcePath("/biz/k11i/xgboost/demo/model/spark/agaricus.txt.test")
+    try {
+      run(sparkSession, tempFileResource)
+    } finally {
+      tempFileResource.close()
+    }
+  }
+
+  def run(sparkSession: SparkSession, tempFileResource: TemporaryFileResource) {
+    val modelPath = tempFileResource.getAsPath("model/gbtree/spark/agaricus.model.spark").toString
+    val testDataPath = tempFileResource.getAsPath("data/agaricus.txt.1.test").toString
 
     val binaryClassifier = XGBoostBinaryClassification.load(modelPath)
       .setRawPredictionCol("rawPrediction")
@@ -36,6 +46,4 @@ object BinaryClassificationExample {
 
     println(s"AUC: $areaUnderROC")
   }
-
-  def getResourcePath(name: String): String = getClass.getResource(name).getPath
 }

xgboost-predictor-examples/src/main/scala/biz/k11i/xgboost/spark/demo/MultiClassificationExample.scala

@@ -1,5 +1,6 @@
 package biz.k11i.xgboost.spark.demo
 
+import biz.k11i.xgboost.TemporaryFileResource
 import biz.k11i.xgboost.spark.model.XGBoostMultiClassification
 import org.apache.spark.SparkConf
 import org.apache.spark.sql.SparkSession
@@ -9,9 +10,18 @@ object MultiClassificationExample {
     val sparkConf = new SparkConf().setAppName("MultiClassificationExample")
       .setMaster("local")
     val sparkSession = SparkSession.builder().config(sparkConf).getOrCreate()
+    val tempFileResource = new TemporaryFileResource
 
-    val modelPath = getResourcePath("/biz/k11i/xgboost/demo/model/spark/iris.model.spark")
-    val testDataPath = getResourcePath("/biz/k11i/xgboost/demo/model/spark/iris.test")
+    try {
+      run(sparkSession, tempFileResource)
+    } finally {
+      tempFileResource.close()
+    }
+  }
+
+  def run(sparkSession: SparkSession, tempFileResource: TemporaryFileResource) {
+    val modelPath = tempFileResource.getAsPath("model/gbtree/spark/iris.model.spark").toString
+    val testDataPath = tempFileResource.getAsPath("data/iris.test").toString
 
     val multiclassClassifier = XGBoostMultiClassification.load(modelPath)
       .setRawPredictionCol("rawPrediction")
@@ -22,6 +32,4 @@ object MultiClassificationExample {
       .select("rawPrediction", "probability", "prediction", "label")
       .show(false)
   }
-
-  def getResourcePath(name: String): String = getClass.getResource(name).getPath
 }

xgboost-predictor-examples/src/main/scala/biz/k11i/xgboost/spark/demo/RegressionExample.scala

@@ -1,5 +1,6 @@
 package biz.k11i.xgboost.spark.demo
 
+import biz.k11i.xgboost.TemporaryFileResource
 import biz.k11i.xgboost.spark.model.XGBoostRegression
 import org.apache.spark.SparkConf
 import org.apache.spark.ml.evaluation.RegressionEvaluator
@@ -10,9 +11,18 @@ object RegressionExample {
     val sparkConf = new SparkConf().setAppName("RegressionExample")
       .setMaster("local")
     val sparkSession = SparkSession.builder().config(sparkConf).getOrCreate()
+    val tempFileResource = new TemporaryFileResource
 
-    val modelPath = getResourcePath("/biz/k11i/xgboost/demo/model/spark/housing.model.spark")
-    val testDataPath = getResourcePath("/biz/k11i/xgboost/demo/model/spark/housing.test")
+    try {
+      run(sparkSession, tempFileResource)
+    } finally {
+      tempFileResource.close()
+    }
+  }
+
+  def run(sparkSession: SparkSession, tempFileResource: TemporaryFileResource) {
+    val modelPath = tempFileResource.getAsPath("model/gbtree/spark/housing.model.spark").toString
+    val testDataPath = tempFileResource.getAsPath("data/housing.test").toString
 
     val regressor = XGBoostRegression.load(modelPath)
     val df = sparkSession.sqlContext.read
@@ -32,6 +42,4 @@ object RegressionExample {
 
     println(s"RMSE: $rmse")
   }
-
-  def getResourcePath(name: String): String = getClass.getResource(name).getPath
 }