add default param checks for CountVectorizerModel

apache · Mar 29, 2018 · ecafd2b · ecafd2b
1 parent a7755fd
commit ecafd2b
Showing 1 changed file with 5 additions and 0 deletions.
diff --git a/python/pyspark/ml/tests.py b/python/pyspark/ml/tests.py
@@ -2096,6 +2096,11 @@ def test_java_params(self):
                     # NOTE: disable check_params_exist until there is parity with Scala API
                     ParamTests.check_params(self, cls(), check_params_exist=False)
 
+        # Additional classes that need explicit construction
+        from pyspark.ml.feature import CountVectorizerModel
+        ParamTests.check_params(self, CountVectorizerModel.from_vocabulary(['a'], 'input'),
+                                check_params_exist=False)
+
 
 def _squared_distance(a, b):
     if isinstance(a, Vector):