[MINOR][ML] Remove unnecessary default value setting for evaluators.

## What changes were proposed in this pull request?
Remove unnecessary default value setting for all evaluators, as we have set them in corresponding _HasXXX_ base classes.

## How was this patch tested?
Existing tests.

Author: Yanbo Liang <ybliang8@gmail.com>

Closes apache#19262 from yanboliang/evaluation.

python/pyspark/ml/evaluation.py

@@ -146,8 +146,7 @@ def __init__(self, rawPredictionCol="rawPrediction", labelCol="label",
         super(BinaryClassificationEvaluator, self).__init__()
         self._java_obj = self._new_java_obj(
             "org.apache.spark.ml.evaluation.BinaryClassificationEvaluator", self.uid)
-        self._setDefault(rawPredictionCol="rawPrediction", labelCol="label",
-                         metricName="areaUnderROC")
+        self._setDefault(metricName="areaUnderROC")
         kwargs = self._input_kwargs
         self._set(**kwargs)
 
@@ -224,8 +223,7 @@ def __init__(self, predictionCol="prediction", labelCol="label",
         super(RegressionEvaluator, self).__init__()
         self._java_obj = self._new_java_obj(
             "org.apache.spark.ml.evaluation.RegressionEvaluator", self.uid)
-        self._setDefault(predictionCol="prediction", labelCol="label",
-                         metricName="rmse")
+        self._setDefault(metricName="rmse")
         kwargs = self._input_kwargs
         self._set(**kwargs)
 
@@ -297,8 +295,7 @@ def __init__(self, predictionCol="prediction", labelCol="label",
         super(MulticlassClassificationEvaluator, self).__init__()
         self._java_obj = self._new_java_obj(
             "org.apache.spark.ml.evaluation.MulticlassClassificationEvaluator", self.uid)
-        self._setDefault(predictionCol="prediction", labelCol="label",
-                         metricName="f1")
+        self._setDefault(metricName="f1")
         kwargs = self._input_kwargs
         self._set(**kwargs)