[SPARK-15771][ML][EXAMPLES] Use 'accuracy' rather than 'precision' in many ML examples

## What changes were proposed in this pull request?
Since [SPARK-15617](https://issues.apache.org/jira/browse/SPARK-15617) deprecated ```precision``` in ```MulticlassClassificationEvaluator```, many ML examples broken.
```python
pyspark.sql.utils.IllegalArgumentException: u'MulticlassClassificationEvaluator_4c3bb1d73d8cc0cedae6 parameter metricName given invalid value precision.'
```
We should use ```accuracy``` to replace ```precision``` in these examples.

## How was this patch tested?
Offline tests.

Author: Yanbo Liang <ybliang8@gmail.com>

Closes #13519 from yanboliang/spark-15771.

examples/src/main/java/org/apache/spark/examples/ml/JavaDecisionTreeClassificationExample.java

@@ -90,7 +90,7 @@ public class JavaDecisionTreeClassificationExample {
     MulticlassClassificationEvaluator evaluator = new MulticlassClassificationEvaluator()
       .setLabelCol("indexedLabel")
       .setPredictionCol("prediction")
-      .setMetricName("precision");
+      .setMetricName("accuracy");
     double accuracy = evaluator.evaluate(predictions);
     System.out.println("Test Error = " + (1.0 - accuracy));
 

examples/src/main/java/org/apache/spark/examples/ml/JavaGradientBoostedTreeClassifierExample.java

@@ -92,7 +92,7 @@ public class JavaGradientBoostedTreeClassifierExample {
     MulticlassClassificationEvaluator evaluator = new MulticlassClassificationEvaluator()
       .setLabelCol("indexedLabel")
       .setPredictionCol("prediction")
-      .setMetricName("precision");
+      .setMetricName("accuracy");
     double accuracy = evaluator.evaluate(predictions);
     System.out.println("Test Error = " + (1.0 - accuracy));
 

examples/src/main/java/org/apache/spark/examples/ml/JavaMultilayerPerceptronClassifierExample.java

@@ -57,12 +57,12 @@ public class JavaMultilayerPerceptronClassifierExample {
       .setMaxIter(100);
     // train the model
     MultilayerPerceptronClassificationModel model = trainer.fit(train);
-    // compute precision on the test set
+    // compute accuracy on the test set
     Dataset<Row> result = model.transform(test);
     Dataset<Row> predictionAndLabels = result.select("prediction", "label");
     MulticlassClassificationEvaluator evaluator = new MulticlassClassificationEvaluator()
-      .setMetricName("precision");
-    System.out.println("Precision = " + evaluator.evaluate(predictionAndLabels));
+      .setMetricName("accuracy");
+    System.out.println("Accuracy = " + evaluator.evaluate(predictionAndLabels));
     // $example off$
 
     spark.stop();

examples/src/main/java/org/apache/spark/examples/ml/JavaNaiveBayesExample.java

@@ -50,12 +50,12 @@ public class JavaNaiveBayesExample {
     NaiveBayes nb = new NaiveBayes();
     // train the model
     NaiveBayesModel model = nb.fit(train);
-    // compute precision on the test set
+    // compute accuracy on the test set
     Dataset<Row> result = model.transform(test);
     Dataset<Row> predictionAndLabels = result.select("prediction", "label");
     MulticlassClassificationEvaluator evaluator = new MulticlassClassificationEvaluator()
-      .setMetricName("precision");
-    System.out.println("Precision = " + evaluator.evaluate(predictionAndLabels));
+      .setMetricName("accuracy");
+    System.out.println("Accuracy = " + evaluator.evaluate(predictionAndLabels));
     // $example off$
 
     spark.stop();

examples/src/main/java/org/apache/spark/examples/ml/JavaOneVsRestExample.java

@@ -71,11 +71,11 @@ public class JavaOneVsRestExample {
 
     // obtain evaluator.
     MulticlassClassificationEvaluator evaluator = new MulticlassClassificationEvaluator()
-            .setMetricName("precision");
+            .setMetricName("accuracy");
 
     // compute the classification error on test data.
-    double precision = evaluator.evaluate(predictions);
-    System.out.println("Test Error : " + (1 - precision));
+    double accuracy = evaluator.evaluate(predictions);
+    System.out.println("Test Error : " + (1 - accuracy));
     // $example off$
 
     spark.stop();

examples/src/main/java/org/apache/spark/examples/ml/JavaRandomForestClassifierExample.java

@@ -88,7 +88,7 @@ public class JavaRandomForestClassifierExample {
     MulticlassClassificationEvaluator evaluator = new MulticlassClassificationEvaluator()
       .setLabelCol("indexedLabel")
       .setPredictionCol("prediction")
-      .setMetricName("precision");
+      .setMetricName("accuracy");
     double accuracy = evaluator.evaluate(predictions);
     System.out.println("Test Error = " + (1.0 - accuracy));
 

examples/src/main/python/ml/decision_tree_classification_example.py

@@ -66,7 +66,7 @@ if __name__ == "__main__":
 
     # Select (prediction, true label) and compute test error
     evaluator = MulticlassClassificationEvaluator(
-        labelCol="indexedLabel", predictionCol="prediction", metricName="precision")
+        labelCol="indexedLabel", predictionCol="prediction", metricName="accuracy")
     accuracy = evaluator.evaluate(predictions)
     print("Test Error = %g " % (1.0 - accuracy))
 

examples/src/main/python/ml/gradient_boosted_tree_classifier_example.py

@@ -66,7 +66,7 @@ if __name__ == "__main__":
 
     # Select (prediction, true label) and compute test error
     evaluator = MulticlassClassificationEvaluator(
-        labelCol="indexedLabel", predictionCol="prediction", metricName="precision")
+        labelCol="indexedLabel", predictionCol="prediction", metricName="accuracy")
     accuracy = evaluator.evaluate(predictions)
     print("Test Error = %g" % (1.0 - accuracy))
 

examples/src/main/python/ml/multilayer_perceptron_classification.py

@@ -43,11 +43,11 @@ if __name__ == "__main__":
     trainer = MultilayerPerceptronClassifier(maxIter=100, layers=layers, blockSize=128, seed=1234)
     # train the model
     model = trainer.fit(train)
-    # compute precision on the test set
+    # compute accuracy on the test set
     result = model.transform(test)
     predictionAndLabels = result.select("prediction", "label")
-    evaluator = MulticlassClassificationEvaluator(metricName="precision")
-    print("Precision:" + str(evaluator.evaluate(predictionAndLabels)))
+    evaluator = MulticlassClassificationEvaluator(metricName="accuracy")
+    print("Accuracy: " + str(evaluator.evaluate(predictionAndLabels)))
     # $example off$
 
     spark.stop()

examples/src/main/python/ml/naive_bayes_example.py

@@ -43,11 +43,11 @@ if __name__ == "__main__":
 
     # train the model
     model = nb.fit(train)
-    # compute precision on the test set
+    # compute accuracy on the test set
     result = model.transform(test)
     predictionAndLabels = result.select("prediction", "label")
-    evaluator = MulticlassClassificationEvaluator(metricName="precision")
-    print("Precision:" + str(evaluator.evaluate(predictionAndLabels)))
+    evaluator = MulticlassClassificationEvaluator(metricName="accuracy")
+    print("Accuracy: " + str(evaluator.evaluate(predictionAndLabels)))
     # $example off$
 
     spark.stop()

examples/src/main/python/ml/one_vs_rest_example.py

@@ -58,11 +58,11 @@ if __name__ == "__main__":
     predictions = ovrModel.transform(test)
 
     # obtain evaluator.
-    evaluator = MulticlassClassificationEvaluator(metricName="precision")
+    evaluator = MulticlassClassificationEvaluator(metricName="accuracy")
 
     # compute the classification error on test data.
-    precision = evaluator.evaluate(predictions)
-    print("Test Error : " + str(1 - precision))
+    accuracy = evaluator.evaluate(predictions)
+    print("Test Error : " + str(1 - accuracy))
     # $example off$
 
     spark.stop()

examples/src/main/python/ml/random_forest_classifier_example.py

@@ -66,7 +66,7 @@ if __name__ == "__main__":
 
     # Select (prediction, true label) and compute test error
     evaluator = MulticlassClassificationEvaluator(
-        labelCol="indexedLabel", predictionCol="prediction", metricName="precision")
+        labelCol="indexedLabel", predictionCol="prediction", metricName="accuracy")
     accuracy = evaluator.evaluate(predictions)
     print("Test Error = %g" % (1.0 - accuracy))
 

examples/src/main/scala/org/apache/spark/examples/ml/DecisionTreeClassificationExample.scala

@@ -81,7 +81,7 @@ object DecisionTreeClassificationExample {
     val evaluator = new MulticlassClassificationEvaluator()
       .setLabelCol("indexedLabel")
       .setPredictionCol("prediction")
-      .setMetricName("precision")
+      .setMetricName("accuracy")
     val accuracy = evaluator.evaluate(predictions)
     println("Test Error = " + (1.0 - accuracy))
 

examples/src/main/scala/org/apache/spark/examples/ml/GradientBoostedTreeClassifierExample.scala

@@ -83,7 +83,7 @@ object GradientBoostedTreeClassifierExample {
     val evaluator = new MulticlassClassificationEvaluator()
       .setLabelCol("indexedLabel")
       .setPredictionCol("prediction")
-      .setMetricName("precision")
+      .setMetricName("accuracy")
     val accuracy = evaluator.evaluate(predictions)
     println("Test Error = " + (1.0 - accuracy))
 

examples/src/main/scala/org/apache/spark/examples/ml/MultilayerPerceptronClassifierExample.scala

@@ -55,12 +55,12 @@ object MultilayerPerceptronClassifierExample {
       .setMaxIter(100)
     // train the model
     val model = trainer.fit(train)
-    // compute precision on the test set
+    // compute accuracy on the test set
     val result = model.transform(test)
     val predictionAndLabels = result.select("prediction", "label")
     val evaluator = new MulticlassClassificationEvaluator()
-      .setMetricName("precision")
-    println("Precision:" + evaluator.evaluate(predictionAndLabels))
+      .setMetricName("accuracy")
+    println("Accuracy: " + evaluator.evaluate(predictionAndLabels))
     // $example off$
 
     spark.stop()

examples/src/main/scala/org/apache/spark/examples/ml/NaiveBayesExample.scala

@@ -49,9 +49,9 @@ object NaiveBayesExample {
     val evaluator = new MulticlassClassificationEvaluator()
       .setLabelCol("label")
       .setPredictionCol("prediction")
-      .setMetricName("precision")
-    val precision = evaluator.evaluate(predictions)
-    println("Precision:" + precision)
+      .setMetricName("accuracy")
+    val accuracy = evaluator.evaluate(predictions)
+    println("Accuracy: " + accuracy)
     // $example off$
 
     spark.stop()

examples/src/main/scala/org/apache/spark/examples/ml/OneVsRestExample.scala

@@ -65,11 +65,11 @@ object OneVsRestExample {
 
     // obtain evaluator.
     val evaluator = new MulticlassClassificationEvaluator()
-      .setMetricName("precision")
+      .setMetricName("accuracy")
 
     // compute the classification error on test data.
-    val precision = evaluator.evaluate(predictions)
-    println(s"Test Error : ${1 - precision}")
+    val accuracy = evaluator.evaluate(predictions)
+    println(s"Test Error : ${1 - accuracy}")
     // $example off$
 
     spark.stop()

examples/src/main/scala/org/apache/spark/examples/ml/RandomForestClassifierExample.scala

@@ -83,7 +83,7 @@ object RandomForestClassifierExample {
     val evaluator = new MulticlassClassificationEvaluator()
       .setLabelCol("indexedLabel")
       .setPredictionCol("prediction")
-      .setMetricName("precision")
+      .setMetricName("accuracy")
     val accuracy = evaluator.evaluate(predictions)
     println("Test Error = " + (1.0 - accuracy))
 

python/pyspark/ml/evaluation.py

@@ -265,7 +265,7 @@ class MulticlassClassificationEvaluator(JavaEvaluator, HasLabelCol, HasPredictio
     """
     metricName = Param(Params._dummy(), "metricName",
                        "metric name in evaluation "
-                       "(f1|precision|recall|weightedPrecision|weightedRecall|accuracy)",
+                       "(f1|weightedPrecision|weightedRecall|accuracy)",
                        typeConverter=TypeConverters.toString)
 
     @keyword_only