MacOS OpenMP and input validation fixes (#1338)

* Extend CI and Nightly timeout to 2h

* Update validation of function input

* Correct KMeans FutureWarning

* Deselect KMeans FutureWarning for 1.4.dev0 sklearn version

* Set SKLEARN_SKIP_OPENMP_TEST in macOS nightly

.ci/pipeline/ci.yml

@@ -60,6 +60,7 @@ jobs:
     displayName: 'PEP 8 check'
 
 - job: Linux
+  timeoutInMinutes: 120
   strategy:
     matrix:
       Python3.7_Sklearn1.0:
@@ -82,6 +83,7 @@ jobs:
   steps:
   - template: build-and-test-lnx.yml
 - job: MacOS
+  timeoutInMinutes: 120
   strategy:
     matrix:
       Python3.7_Sklearn1.0:
@@ -104,6 +106,7 @@ jobs:
   steps:
   - template: build-and-test-mac.yml
 - job: Windows
+  timeoutInMinutes: 120
   strategy:
     matrix:
       Python3.7_Sklearn1.0:

.ci/pipeline/nightly.yml

@@ -82,6 +82,7 @@ jobs:
     displayName: 'Run jupyter notebook demo'
 
 - job: LinuxNightly
+  timeoutInMinutes: 120
   strategy:
     matrix:
       Python3.10_SklearnMain:
@@ -92,16 +93,20 @@ jobs:
   steps:
   - template: build-and-test-lnx.yml
 - job: MacOSNightly
+  timeoutInMinutes: 120
   strategy:
     matrix:
       Python3.10_SklearnMain:
         PYTHON_VERSION: '3.10'
         SKLEARN_VERSION: 'main'
   pool:
     vmImage: 'macos-12'
+  variables:
+    SKLEARN_SKIP_OPENMP_TEST: 'true'
   steps:
   - template: build-and-test-mac.yml
 - job: WindowsNightly
+  timeoutInMinutes: 120
   strategy:
     matrix:
       Python3.10_SklearnMain:

daal4py/sklearn/cluster/_k_means_0_23.py

@@ -180,7 +180,7 @@ def is_string(s, target_str):
 
     default_n_init = 10
     if n_init in ['auto', 'warn']:
-        if n_init == "warn" and not sklearn_check_version('1.2'):
+        if n_init == "warn" and sklearn_check_version('1.2'):
             warnings.warn(
                 "The default value of `n_init` will change from "
                 f"{default_n_init} to 'auto' in 1.4. Set the value of `n_init`"

daal4py/sklearn/metrics/_ranking.py

@@ -183,6 +183,8 @@ def roc_auc_score(
 
 
 if sklearn_check_version('1.3'):
+    validation_kwargs = {'prefer_skip_nested_validation': True} \
+        if sklearn_check_version('1.4') else {}
     roc_auc_score = validate_params(
         {
             "y_true": ["array-like"],
@@ -192,5 +194,5 @@ def roc_auc_score(
             "max_fpr": [Interval(Real, 0.0, 1, closed="right"), None],
             "multi_class": [StrOptions({"raise", "ovr", "ovo"})],
             "labels": ["array-like", None],
-        }
+        }, **validation_kwargs
     )(roc_auc_score)

daal4py/sklearn/model_selection/_split.py

@@ -258,6 +258,8 @@ def train_test_split(*arrays, **options):
 
 
 if sklearn_check_version('1.3'):
+    validation_kwargs = {'prefer_skip_nested_validation': True} \
+        if sklearn_check_version('1.4') else {}
     train_test_split = validate_params({
         "test_size": [
             Interval(RealNotInt, 0, 1, closed="neither"),
@@ -272,4 +274,4 @@ def train_test_split(*arrays, **options):
         "random_state": ["random_state"],
         "shuffle": ["boolean"],
         "stratify": ["array-like", None],
-    })(train_test_split)
+    }, **validation_kwargs)(train_test_split)

deselected_tests.yaml

@@ -25,6 +25,9 @@
 #  will exclude deselection in versions 0.18.1, and 0.18.2 only.
 
 deselected_tests:
+  # test for KMeans FutureWarning is not removed from sklearn tests suit yet
+  - cluster/tests/test_k_means.py::test_change_n_init_future_warning[KMeans-10] ==1.4.dev0
+
   # Non-critical, but there are significant numerical differences in doctest results
   - pipeline.py::sklearn.pipeline.FeatureUnion
   - ensemble/_forest.py::sklearn.ensemble._forest.RandomForestRegressor