Addressed follow up comments

python/pyspark/sql/dataframe.py

@@ -437,6 +437,10 @@ def sample(self, withReplacement, fraction, seed=None):
     def randomSplit(self, weights, seed=None):
         """Randomly splits this :class:`DataFrame` with the provided weights.
 
+        :param weights: list of doubles with which to split the DataFrame into. Weights will be
+            normalized if they don't sum up to 1.0.
+        :param seed: The seed for sampling.
+
         >>> splits = df4.randomSplit([1.0, 2.0], 24)
         >>> splits[0].count()
         1
@@ -445,7 +449,8 @@ def randomSplit(self, weights, seed=None):
         3
         """
         for w in weights:
-            assert w >= 0.0, "Negative weight value: %s" % w
+            if w < 0.0:
+                raise ValueError("Weights must be positive. Found weight value: %s" % w)
         seed = seed if seed is not None else random.randint(0, sys.maxsize)
         rdd_array = self._jdf.randomSplit(_to_seq(self.sql_ctx._sc, weights), long(seed))
         return [DataFrame(rdd, self.sql_ctx) for rdd in rdd_array]

sql/core/src/main/scala/org/apache/spark/sql/DataFrame.scala

@@ -752,7 +752,7 @@ class DataFrame private[sql](
    * @param seed Seed for sampling.
    * @group dfops
    */
-  def randomSplit(weights: List[Double], seed: Long): Array[DataFrame] = {
+  private[spark] def randomSplit(weights: List[Double], seed: Long): Array[DataFrame] = {
     randomSplit(weights.toArray, seed)
   }