Skip to content

Adding support for scikit-learn > 0.22 #936

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
PGijsbers merged 5 commits into from
Aug 3, 2020
Merged

Adding support for scikit-learn > 0.22 #936

Show file tree
Hide file tree
Changes from 1 commit
Commits
Show all changes
5 commits
Select commit Hold shift + click to select a range
46c25fd
Preliminary changes
Neeratyoy Jul 27, 2020
9ed6a09
Merge branch 'develop' into sklearn-0.22+support
Neeratyoy Jul 27, 2020
3f736b7
Updating unit tests for sklearn 0.22 and above
Neeratyoy Jul 28, 2020
dcf2d22
Triggering sklearn tests + fixes
Neeratyoy Jul 29, 2020
023ac40
Refactoring to inspect.signature in extensions
Neeratyoy Jul 31, 2020
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
Next Next commit
Preliminary changes
  • Loading branch information
@Neeratyoy
Neeratyoy committed Jul 27, 2020
commit 46c25fdfdfac32eb44d7bcd1683d31176a7e75f9
77 changes: 55 additions & 22 deletions tests/test_flows/test_flow.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -305,15 +305,27 @@ def test_publish_error(self, api_call_mock, flow_exists_mock, get_flow_mock):
"collected from {}: {}".format(__file__.split("/")[-1], flow.flow_id)
)

fixture = (
"The flow on the server is inconsistent with the local flow. "
"The server flow ID is 1. Please check manually and remove "
"the flow if necessary! Error is:\n"
"'Flow sklearn.ensemble.forest.RandomForestClassifier: "
"values for attribute 'name' differ: "
"'sklearn.ensemble.forest.RandomForestClassifier'"
"\nvs\n'sklearn.ensemble.forest.RandomForestClassifie'.'"
)
if LooseVersion(sklearn.__version__) < "0.22":
fixture = (
"The flow on the server is inconsistent with the local flow. "
"The server flow ID is 1. Please check manually and remove "
"the flow if necessary! Error is:\n"
"'Flow sklearn.ensemble.forest.RandomForestClassifier: "
"values for attribute 'name' differ: "
"'sklearn.ensemble.forest.RandomForestClassifier'"
"\nvs\n'sklearn.ensemble.forest.RandomForestClassifie'.'"
)
else:
# sklearn.ensemble.forest -> sklearn.ensemble._forest
fixture = (
"The flow on the server is inconsistent with the local flow. "
"The server flow ID is 1. Please check manually and remove "
"the flow if necessary! Error is:\n"
"'Flow sklearn.ensemble._forest.RandomForestClassifier: "
"values for attribute 'name' differ: "
"'sklearn.ensemble._forest.RandomForestClassifier'"
"\nvs\n'sklearn.ensemble._forest.RandomForestClassifie'.'"
)

self.assertEqual(context_manager.exception.args[0], fixture)
self.assertEqual(get_flow_mock.call_count, 2)
Expand Down Expand Up @@ -463,19 +475,40 @@ def test_sklearn_to_upload_to_flow(self):

# OneHotEncoder was moved to _encoders module in 0.20
module_name_encoder = "_encoders" if LooseVersion(sklearn.__version__) >= "0.20" else "data"
fixture_name = (
"%ssklearn.model_selection._search.RandomizedSearchCV("
"estimator=sklearn.pipeline.Pipeline("
"ohe=sklearn.preprocessing.%s.OneHotEncoder,"
"scaler=sklearn.preprocessing.data.StandardScaler,"
"fu=sklearn.pipeline.FeatureUnion("
"pca=sklearn.decomposition.truncated_svd.TruncatedSVD,"
"fs="
"sklearn.feature_selection.univariate_selection.SelectPercentile),"
"boosting=sklearn.ensemble.weight_boosting.AdaBoostClassifier("
"base_estimator=sklearn.tree.tree.DecisionTreeClassifier)))"
% (sentinel, module_name_encoder)
)
if LooseVersion(sklearn.__version__) < "0.22":
fixture_name = (
"%ssklearn.model_selection._search.RandomizedSearchCV("
"estimator=sklearn.pipeline.Pipeline("
"ohe=sklearn.preprocessing.%s.OneHotEncoder,"
"scaler=sklearn.preprocessing.data.StandardScaler,"
"fu=sklearn.pipeline.FeatureUnion("
"pca=sklearn.decomposition.truncated_svd.TruncatedSVD,"
"fs="
"sklearn.feature_selection.univariate_selection.SelectPercentile),"
"boosting=sklearn.ensemble.weight_boosting.AdaBoostClassifier("
"base_estimator=sklearn.tree.tree.DecisionTreeClassifier)))"
% (sentinel, module_name_encoder)
)
else:
# sklearn.sklearn.preprocessing.data -> sklearn.sklearn.preprocessing._data
# sklearn.sklearn.decomposition.truncated_svd -> sklearn.decomposition._truncated_svd
# sklearn.feature_selection.univariate_selection ->
# sklearn.feature_selection._univariate_selection
# sklearn.ensemble.weight_boosting -> sklearn.ensemble._weight_boosting
# sklearn.tree.tree.DecisionTree... -> sklearn.tree._classes.DecisionTree...
fixture_name = (
"%ssklearn.model_selection._search.RandomizedSearchCV("
"estimator=sklearn.pipeline.Pipeline("
"ohe=sklearn.preprocessing.%s.OneHotEncoder,"
"scaler=sklearn.preprocessing._data.StandardScaler,"
"fu=sklearn.pipeline.FeatureUnion("
"pca=sklearn.decomposition._truncated_svd.TruncatedSVD,"
"fs="
"sklearn.feature_selection._univariate_selection.SelectPercentile),"
"boosting=sklearn.ensemble._weight_boosting.AdaBoostClassifier("
"base_estimator=sklearn.tree._classes.DecisionTreeClassifier)))"
% (sentinel, module_name_encoder)
)
self.assertEqual(new_flow.name, fixture_name)
new_flow.model.fit(X, y)

Expand Down
10 changes: 8 additions & 2 deletions tests/test_runs/test_run_functions.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -199,8 +199,11 @@ def _perform_run(
classes_without_random_state = [
"sklearn.model_selection._search.GridSearchCV",
"sklearn.pipeline.Pipeline",
"sklearn.linear_model.base.LinearRegression",
]
if LooseVersion(sklearn.__version__) < "0.22":
classes_without_random_state.append("sklearn.linear_model.base.LinearRegression")
else:
classes_without_random_state.append("sklearn.linear_model._base.LinearRegression")

def _remove_random_state(flow):
if "random_state" in flow.parameters:
Expand Down Expand Up @@ -779,10 +782,13 @@ def _test_local_evaluations(self, run):
(sklearn.metrics.cohen_kappa_score, {"weights": None}),
(sklearn.metrics.roc_auc_score, {}),
(sklearn.metrics.average_precision_score, {}),
(sklearn.metrics.jaccard_similarity_score, {}),
(sklearn.metrics.precision_score, {"average": "macro"}),
(sklearn.metrics.brier_score_loss, {}),
]
if LooseVersion(sklearn.__version__) < "0.23":
tests.append((sklearn.metrics.jaccard_similarity_score, {}))
else:
tests.append((sklearn.metrics.jaccard_score, {}))
for test_idx, test in enumerate(tests):
alt_scores = run.get_metric_fn(sklearn_fn=test[0], kwargs=test[1],)
self.assertEqual(len(alt_scores), 10)
Expand Down