Remove random_state arg from get_pipelines in AutoMLSearch

docs/source/release_notes.rst

@@ -7,6 +7,7 @@ Release Notes
         * Added the ability to see the linear coefficients of features in linear models terms :pr:`1738`
         * Added support for ``scikit-learn`` ``v0.24.0`` :pr:`1733`
     * Fixes
+        * Removed ``random_state`` arg from ``get_pipelines`` in ``AutoMLSearch`` :pr:`1719`
     * Changes
         * Updated components and pipelines to return ``Woodwork`` data structures :pr:`1668`
     * Documentation Changes

evalml/automl/automl_search.py

@@ -840,13 +840,12 @@ def _evaluate_pipelines(self, current_pipeline_batch, baseline=False, search_ite
 
         return current_batch_pipeline_scores
 
-    def get_pipeline(self, pipeline_id, random_state=0):
+    def get_pipeline(self, pipeline_id):
         """Given the ID of a pipeline training result, returns an untrained instance of the specified pipeline
         initialized with the parameters used to train that pipeline during automl search.
 
         Arguments:
             pipeline_id (int): pipeline to retrieve
-            random_state (int): Seed for the random number generator. Defaults to 0.
 
         Returns:
             PipelineBase: untrained pipeline instance associated with the provided ID
@@ -862,7 +861,7 @@ def get_pipeline(self, pipeline_id, random_state=0):
         pipeline.custom_hyperparameters = pipeline_class.custom_hyperparameters
         pipeline.custom_name = pipeline_class.name
         pipeline.component_graph = pipeline_class.component_graph
-        return pipeline(parameters, random_state=random_state)
+        return pipeline(parameters, random_state=self.random_state)
 
     def describe_pipeline(self, pipeline_id, return_dict=False):
         """Describe a pipeline

evalml/tests/automl_tests/test_automl.py

@@ -2165,3 +2165,16 @@ def test_automl_pipeline_params_kwargs(mock_fit, mock_score, X_y_multi):
         if 'Decision Tree Classifier' in row['parameters']:
             assert 0.1 < row['parameters']['Decision Tree Classifier']['ccp_alpha'] < 0.5
             assert row['parameters']['Decision Tree Classifier']['max_depth'] == 1
+
+
+@pytest.mark.parametrize("random_state", [0, 1, 9])
+@patch('evalml.pipelines.MulticlassClassificationPipeline.score')
+@patch('evalml.pipelines.MulticlassClassificationPipeline.fit')
+def test_automl_pipeline_random_state(mock_fit, mock_score, random_state, X_y_multi):
+    X, y = X_y_multi
+    automl = AutoMLSearch(X_train=X, y_train=y, problem_type='multiclass', random_state=random_state, n_jobs=1)
+    automl.search()
+
+    for i, row in automl.rankings.iterrows():
+        if 'Base' not in list(row['parameters'].keys())[0]:
+            assert automl.get_pipeline(row['id']).random_state == random_state