Fix python unit test and add document.

Author: viirya

mllib/src/main/scala/org/apache/spark/ml/evaluation/RegressionEvaluator.scala

@@ -37,6 +37,10 @@ final class RegressionEvaluator(override val uid: String)
 
   /**
    * param for metric name in evaluation (supports `"rmse"` (default), `"mse"`, `"r2"`, and `"mae"`)
+   *
+   * Because we will maximize evaluation value (ref: `CrossValidator`),
+   * when we evaluate a metric that is needed to minimize (e.g., `"rmse"`, `"mse"`, `"mae"`),
+   * we take and output the negative of this metric.
    * @group param
    */
   val metricName: Param[String] = {

python/pyspark/ml/evaluation.py

@@ -160,13 +160,15 @@ class RegressionEvaluator(JavaEvaluator, HasLabelCol, HasPredictionCol):
     ...
     >>> evaluator = RegressionEvaluator(predictionCol="raw")
     >>> evaluator.evaluate(dataset)
-    2.842...
+    -2.842...
     >>> evaluator.evaluate(dataset, {evaluator.metricName: "r2"})
     0.993...
     >>> evaluator.evaluate(dataset, {evaluator.metricName: "mae"})
-    2.649...
+    -2.649...
     """
-    # a placeholder to make it appear in the generated doc
+    # Because we will maximize evaluation value (ref: `CrossValidator`),
+    # when we evaluate a metric that is needed to minimize (e.g., `"rmse"`, `"mse"`, `"mae"`),
+    # we take and output the negative of this metric.
     metricName = Param(Params._dummy(), "metricName",
                        "metric name in evaluation (mse|rmse|r2|mae)")