Undo whitespace changes to files not touched otherwise

Author: PGijsbers

tests/test_flows/test_flow.py

@@ -5,7 +5,6 @@
 import copy
 import hashlib
 import re
-import os
 import time
 from packaging.version import Version
 from unittest import mock
@@ -34,6 +33,7 @@
 from openml.testing import SimpleImputer, TestBase
 
 
+
 class TestFlow(TestBase):
     _multiprocess_can_split_ = True
 
@@ -179,6 +179,7 @@ def test_to_xml_from_xml(self):
         # Would raise exception if they are not legal
         openml.flows.functions.assert_flows_equal(new_flow, flow)
         assert new_flow is not flow
+
     @pytest.mark.sklearn()
     @pytest.mark.test_server()
     def test_publish_flow(self):
@@ -220,6 +221,7 @@ def test_publish_existing_flow(self, flow_exists_mock):
         TestBase.logger.info(
             f"collected from {__file__.split('/')[-1]}: {flow.flow_id}",
         )
+
     @pytest.mark.sklearn()
     @pytest.mark.test_server()
     def test_publish_flow_with_similar_components(self):
@@ -270,6 +272,7 @@ def test_publish_flow_with_similar_components(self):
         flow3.publish()
         TestBase._mark_entity_for_removal("flow", flow3.flow_id, flow3.name)
         TestBase.logger.info(f"collected from {__file__.split('/')[-1]}: {flow3.flow_id}")
+
     @pytest.mark.sklearn()
     @pytest.mark.test_server()
     def test_semi_legal_flow(self):
@@ -379,6 +382,7 @@ def get_sentinel():
 
         flow_id = openml.flows.flow_exists(name, version)
         assert not flow_id
+
     @pytest.mark.sklearn()
     @pytest.mark.test_server()
     def test_existing_flow_exists(self):
@@ -419,6 +423,7 @@ def test_existing_flow_exists(self):
                 flow.external_version,
             )
             assert downloaded_flow_id == flow.flow_id
+
     @pytest.mark.sklearn()
     @pytest.mark.test_server()
     def test_sklearn_to_upload_to_flow(self):

tests/test_flows/test_flow_functions.py

@@ -12,7 +12,6 @@
 from unittest import mock
 from unittest.mock import patch
 
-import os
 import pandas as pd
 import pytest
 import requests
@@ -309,6 +308,7 @@ def test_get_flow1(self):
         self.use_production_server()
         flow = openml.flows.get_flow(1)
         assert flow.external_version is None
+
     @pytest.mark.sklearn()
     @pytest.mark.test_server()
     def test_get_flow_reinstantiate_model(self):
@@ -391,6 +391,7 @@ def test_get_flow_reinstantiate_flow_not_strict_pre_023(self):
         flow = openml.flows.get_flow(flow_id=8175, reinstantiate=True, strict_version=False)
         assert flow.flow_id is None
         assert "sklearn==0.19.1" not in flow.dependencies
+
     @pytest.mark.sklearn()
     @pytest.mark.test_server()
     def test_get_flow_id(self):

tests/test_openml/test_api_calls.py

@@ -7,7 +7,6 @@
 
 import minio
 import pytest
-import os
 
 import openml
 from openml.config import ConfigurationForExamples
@@ -20,6 +19,7 @@ class TestConfig(openml.testing.TestBase):
     def test_too_long_uri(self):
         with pytest.raises(openml.exceptions.OpenMLServerError, match="URI too long!"):
             openml.datasets.list_datasets(data_id=list(range(10000)))
+
     @unittest.mock.patch("time.sleep")
     @unittest.mock.patch("requests.Session")
     @pytest.mark.test_server()

tests/test_runs/test_run.py

@@ -117,6 +117,7 @@ def _check_array(array, type_):
             np.testing.assert_array_equal(string_part, string_part_prime)
         else:
             assert run_prime_trace_content is None
+
     @pytest.mark.sklearn()
     @pytest.mark.test_server()
     def test_to_from_filesystem_vanilla(self):
@@ -151,6 +152,7 @@ def test_to_from_filesystem_vanilla(self):
         TestBase.logger.info(
             f"collected from {__file__.split('/')[-1]}: {run_prime.run_id}",
         )
+
     @pytest.mark.sklearn()
     @pytest.mark.flaky()
     @pytest.mark.test_server()
@@ -186,6 +188,7 @@ def test_to_from_filesystem_search(self):
         TestBase.logger.info(
             f"collected from {__file__.split('/')[-1]}: {run_prime.run_id}",
         )
+
     @pytest.mark.sklearn()
     @pytest.mark.test_server()
     def test_to_from_filesystem_no_model(self):
@@ -291,6 +294,7 @@ def assert_run_prediction_data(task, run, model):
             # Assert correctness
             assert_method(y_pred, saved_y_pred)
             assert_method(y_test, saved_y_test)
+
     @pytest.mark.sklearn()
     @pytest.mark.test_server()
     def test_publish_with_local_loaded_flow(self):
@@ -334,6 +338,7 @@ def test_publish_with_local_loaded_flow(self):
             # make sure the flow is published as part of publishing the run.
             assert openml.flows.flow_exists(flow.name, flow.external_version)
             openml.runs.get_run(loaded_run.run_id)
+
     @pytest.mark.sklearn()
     @pytest.mark.test_server()
     def test_offline_and_online_run_identical(self):

tests/test_runs/test_run_functions.py

@@ -396,6 +396,7 @@ def _check_sample_evaluations(
                                 # Windows seems to get an eval-time of 0 sometimes.
                                 assert evaluation > 0
                             assert evaluation < max_time_allowed
+
     @pytest.mark.sklearn()
     @pytest.mark.test_server()
     def test_run_regression_on_classif_task(self):
@@ -412,6 +413,7 @@ def test_run_regression_on_classif_task(self):
                 model=clf,
                 task=task,
             )
+
     @pytest.mark.sklearn()
     @pytest.mark.test_server()
     def test_check_erronous_sklearn_flow_fails(self):
@@ -624,6 +626,7 @@ def _run_and_upload_regression(
             task_type=task_type,
             sentinel=sentinel,
         )
+
     @pytest.mark.sklearn()
     @pytest.mark.test_server()
     def test_run_and_upload_logistic_regression(self):
@@ -632,6 +635,7 @@ def test_run_and_upload_logistic_regression(self):
         n_missing_vals = self.TEST_SERVER_TASK_SIMPLE["n_missing_vals"]
         n_test_obs = self.TEST_SERVER_TASK_SIMPLE["n_test_obs"]
         self._run_and_upload_classification(lr, task_id, n_missing_vals, n_test_obs, "62501")
+
     @pytest.mark.sklearn()
     @pytest.mark.test_server()
     def test_run_and_upload_linear_regression(self):
@@ -662,6 +666,7 @@ def test_run_and_upload_linear_regression(self):
         n_missing_vals = self.TEST_SERVER_TASK_REGRESSION["n_missing_vals"]
         n_test_obs = self.TEST_SERVER_TASK_REGRESSION["n_test_obs"]
         self._run_and_upload_regression(lr, task_id, n_missing_vals, n_test_obs, "62501")
+
     @pytest.mark.sklearn()
     @pytest.mark.test_server()
     def test_run_and_upload_pipeline_dummy_pipeline(self):
@@ -675,6 +680,7 @@ def test_run_and_upload_pipeline_dummy_pipeline(self):
         n_missing_vals = self.TEST_SERVER_TASK_SIMPLE["n_missing_vals"]
         n_test_obs = self.TEST_SERVER_TASK_SIMPLE["n_test_obs"]
         self._run_and_upload_classification(pipeline1, task_id, n_missing_vals, n_test_obs, "62501")
+
     @pytest.mark.sklearn()
     @unittest.skipIf(
         Version(sklearn.__version__) < Version("0.20"),
@@ -791,6 +797,7 @@ def test_run_and_upload_knn_pipeline(self, warnings_mock):
             if _warnings[0][0] == warning_msg:
                 call_count += 1
         assert call_count == 3
+
     @pytest.mark.sklearn()
     @pytest.mark.test_server()
     def test_run_and_upload_gridsearch(self):
@@ -813,6 +820,7 @@ def test_run_and_upload_gridsearch(self):
             flow_expected_rsv="62501",
         )
         assert len(run.trace.trace_iterations) == 9
+
     @pytest.mark.sklearn()
     @pytest.mark.test_server()
     def test_run_and_upload_randomsearch(self):
@@ -845,6 +853,7 @@ def test_run_and_upload_randomsearch(self):
         assert len(run.trace.trace_iterations) == 5
         trace = openml.runs.get_run_trace(run.run_id)
         assert len(trace.trace_iterations) == 5
+
     @pytest.mark.sklearn()
     @pytest.mark.test_server()
     def test_run_and_upload_maskedarrays(self):
@@ -872,6 +881,7 @@ def test_run_and_upload_maskedarrays(self):
         )
 
     ##########################################################################
+
     @pytest.mark.sklearn()
     @pytest.mark.test_server()
     def test_learning_curve_task_1(self):
@@ -896,6 +906,7 @@ def test_learning_curve_task_1(self):
             flow_expected_rsv="62501",
         )
         self._check_sample_evaluations(run.sample_evaluations, num_repeats, num_folds, num_samples)
+
     @pytest.mark.sklearn()
     @pytest.mark.test_server()
     def test_learning_curve_task_2(self):
@@ -932,6 +943,7 @@ def test_learning_curve_task_2(self):
             flow_expected_rsv="62501",
         )
         self._check_sample_evaluations(run.sample_evaluations, num_repeats, num_folds, num_samples)
+
     @pytest.mark.sklearn()
     @unittest.skipIf(
         Version(sklearn.__version__) < Version("0.21"),
@@ -1010,6 +1022,7 @@ def _test_local_evaluations(self, run):
             for idx in range(len(alt_scores)):
                 assert alt_scores[idx] >= 0
                 assert alt_scores[idx] <= 1
+
     @pytest.mark.sklearn()
     @pytest.mark.test_server()
     def test_local_run_swapped_parameter_order_model(self):
@@ -1025,6 +1038,7 @@ def test_local_run_swapped_parameter_order_model(self):
         )
 
         self._test_local_evaluations(run)
+
     @pytest.mark.sklearn()
     @unittest.skipIf(
         Version(sklearn.__version__) < Version("0.20"),
@@ -1053,6 +1067,7 @@ def test_local_run_swapped_parameter_order_flow(self):
         )
 
         self._test_local_evaluations(run)
+
     @pytest.mark.sklearn()
     @unittest.skipIf(
         Version(sklearn.__version__) < Version("0.20"),
@@ -1090,6 +1105,7 @@ def test_online_run_metric_score(self):
         run = openml.runs.get_run(9864498)
 
         self._test_local_evaluations(run)
+
     @pytest.mark.sklearn()
     @unittest.skipIf(
         Version(sklearn.__version__) < Version("0.20"),
@@ -1151,6 +1167,7 @@ def test_initialize_model_from_run(self):
 
         assert flowS.components["Imputer"].parameters["strategy"] == '"most_frequent"'
         assert flowS.components["VarianceThreshold"].parameters["threshold"] == "0.05"
+
     @pytest.mark.sklearn()
     @unittest.skipIf(
         Version(sklearn.__version__) < Version("0.20"),
@@ -1210,6 +1227,7 @@ def test__run_exists(self):
             assert setup_exists > 0, "Server says setup of run does not exist."
             run_ids = run_exists(task.task_id, setup_exists)
             assert run_ids, (run_ids, clf)
+
     @pytest.mark.sklearn()
     @pytest.mark.test_server()
     def test_run_with_illegal_flow_id(self):
@@ -1229,6 +1247,7 @@ def test_run_with_illegal_flow_id(self):
                 flow=flow,
                 avoid_duplicate_runs=True,
             )
+
     @pytest.mark.sklearn()
     @pytest.mark.test_server()
     def test_run_with_illegal_flow_id_after_load(self):
@@ -1286,6 +1305,7 @@ def test_run_with_illegal_flow_id_1(self):
                 flow=flow_new,
                 avoid_duplicate_runs=True,
             )
+
     @pytest.mark.sklearn()
     @pytest.mark.test_server()
     def test_run_with_illegal_flow_id_1_after_load(self):
@@ -1324,6 +1344,7 @@ def test_run_with_illegal_flow_id_1_after_load(self):
             expected_message_regex,
             loaded_run.publish,
         )
+
     @pytest.mark.sklearn()
     @unittest.skipIf(
         Version(sklearn.__version__) < Version("0.20"),
@@ -1553,6 +1574,7 @@ def test_get_runs_list_by_tag(self):
         # Don't remove the size restriction: this query is too expensive without
         runs = openml.runs.list_runs(tag="curves", size=2)
         assert len(runs) >= 1
+
     @pytest.mark.sklearn()
     @unittest.skipIf(
         Version(sklearn.__version__) < Version("0.20"),
@@ -1589,6 +1611,7 @@ def test_run_on_dataset_with_missing_labels_dataframe(self):
         for row in data_content:
             # repeat, fold, row_id, 6 confidences, prediction and correct label
             assert len(row) == 12
+
     @pytest.mark.sklearn()
     @unittest.skipIf(
         Version(sklearn.__version__) < Version("0.20"),
@@ -1642,6 +1665,7 @@ def test_get_uncached_run(self):
         openml.config.set_root_cache_directory(self.static_cache_dir)
         with pytest.raises(openml.exceptions.OpenMLCacheException):
             openml.runs.functions._get_cached_run(10)
+
     @pytest.mark.sklearn()
     @pytest.mark.test_server()
     def test_run_flow_on_task_downloaded_flow(self):
@@ -1741,6 +1765,9 @@ def test_format_prediction_task_regression(self):
         ignored_input = [0] * 5
         res = format_prediction(regression, *ignored_input)
         self.assertListEqual(res, [0] * 5)
+
+
+
     @unittest.skipIf(
         Version(sklearn.__version__) < Version("0.20"),
         reason="SimpleImputer doesn't handle mixed type DataFrame as input",
@@ -1836,6 +1863,8 @@ def test_delete_unknown_run(mock_delete, test_files_directory, test_api_key):
     run_url = f"{openml.config.TEST_SERVER_URL}/api/v1/xml/run/9999999"
     assert run_url == mock_delete.call_args.args[0]
     assert test_api_key == mock_delete.call_args.kwargs.get("params", {}).get("api_key")
+
+
 @pytest.mark.sklearn()
 @unittest.skipIf(
     Version(sklearn.__version__) < Version("0.21"),
@@ -1916,6 +1945,8 @@ def test__run_task_get_arffcontent_2(parallel_mock):
         decimal=2,
         err_msg="Observed performance scores deviate from expected ones.",
     )
+
+
 @pytest.mark.sklearn()
 @unittest.skipIf(
     Version(sklearn.__version__) < Version("0.21"),
@@ -1985,7 +2016,6 @@ def test_joblib_backends(parallel_mock, n_jobs, backend, call_count):
         n_jobs=n_jobs,
     )
     from openml_sklearn import SklearnExtension
-
     extension = SklearnExtension()
     with parallel_backend(backend, n_jobs=n_jobs):
         res = openml.runs.functions._run_task_get_arffcontent(
@@ -2002,4 +2032,4 @@ def test_joblib_backends(parallel_mock, n_jobs, backend, call_count):
     # *_time_millis_* not recorded when n_jobs = -1
     assert len(res[2]["predictive_accuracy"][0]) == 10
     assert len(res[3]["predictive_accuracy"][0]) == 10
-    assert parallel_mock.call_count == call_count
+    assert parallel_mock.call_count == call_count