[SPARK-13646][MLLIB] QuantileDiscretizer counts dataset twice in get…

## What changes were proposed in this pull request?

It avoids counting the dataframe twice.

Author: Abou Haydar Elias <abouhaydar.elias@gmail.com>
Author: Elie A <abouhaydar.elias@gmail.com>

Closes #11491 from eliasah/quantile-discretizer-patch.

mllib/src/main/scala/org/apache/spark/ml/feature/QuantileDiscretizer.scala

@@ -118,7 +118,7 @@ object QuantileDiscretizer extends DefaultParamsReadable[QuantileDiscretizer] wi
     require(totalSamples > 0,
       "QuantileDiscretizer requires non-empty input dataset but was given an empty input.")
     val requiredSamples = math.max(numBins * numBins, minSamplesRequired)
-    val fraction = math.min(requiredSamples.toDouble / dataset.count(), 1.0)
+    val fraction = math.min(requiredSamples.toDouble / totalSamples, 1.0)
     dataset.sample(withReplacement = false, fraction, new XORShiftRandom(seed).nextInt()).collect()
   }