[enhancement] conform to sklearn1.6 validate_data API change (#…

…2048)

* initial commit

* temporary

* temp

* first attempt

* tracing now available

* latest ideas

* add skip

* updates

* first light

* save progress

* add validate_data test

* continued fixes

* fixes

* introduce necessary sklearn 1.6 changes

* formatting and swap to conformance

* add SPECIAL_INSTANCES

* formatting

* fix svm test failure

* add docs, namedtuple

* move back to basic python datatypes

* remove daal4py at source

* remove old code

* fix import

* swap to else

onedal/svm/svm.py

@@ -286,7 +286,7 @@ def _ovr_decision_function(self, predictions, confidences, n_classes):
     def _decision_function(self, X, module, queue):
         _check_is_fitted(self)
         X = _check_array(
-            X, dtype=[np.float64, np.float32], force_all_finite=False, accept_sparse="csr"
+            X, dtype=[np.float64, np.float32], force_all_finite=True, accept_sparse="csr"
         )
         _check_n_features(self, X, False)
 

sklearnex/basic_statistics/basic_statistics.py

@@ -28,6 +28,11 @@
 from .._device_offload import dispatch
 from .._utils import PatchingConditionsChain
 
+if sklearn_check_version("1.6"):
+    from sklearn.utils.validation import validate_data
+else:
+    validate_data = BaseEstimator._validate_data
+
 if sklearn_check_version("1.2"):
     from sklearn.utils._param_validation import StrOptions
 
@@ -145,7 +150,7 @@ def _onedal_fit(self, X, sample_weight=None, queue=None):
             self._validate_params()
 
         if sklearn_check_version("1.0"):
-            X = self._validate_data(X, dtype=[np.float64, np.float32], ensure_2d=False)
+            X = validate_data(self, X, dtype=[np.float64, np.float32], ensure_2d=False)
         else:
             X = check_array(X, dtype=[np.float64, np.float32])
 

sklearnex/basic_statistics/incremental_basic_statistics.py

@@ -34,6 +34,11 @@
 import numbers
 import warnings
 
+if sklearn_check_version("1.6"):
+    from sklearn.utils.validation import validate_data
+else:
+    validate_data = BaseEstimator._validate_data
+
 
 @control_n_jobs(decorated_methods=["partial_fit", "_onedal_finalize_fit"])
 class IncrementalBasicStatistics(BaseEstimator):
@@ -145,7 +150,8 @@ def _onedal_partial_fit(self, X, sample_weight=None, queue=None):
         first_pass = not hasattr(self, "n_samples_seen_") or self.n_samples_seen_ == 0
 
         if sklearn_check_version("1.0"):
-            X = self._validate_data(
+            X = validate_data(
+                self,
                 X,
                 dtype=[np.float64, np.float32],
                 reset=first_pass,
@@ -180,7 +186,7 @@ def _onedal_fit(self, X, sample_weight=None, queue=None):
             self._validate_params()
 
         if sklearn_check_version("1.0"):
-            X = self._validate_data(X, dtype=[np.float64, np.float32])
+            X = validate_data(self, X, dtype=[np.float64, np.float32])
         else:
             X = check_array(X, dtype=[np.float64, np.float32])
 

sklearnex/cluster/dbscan.py

@@ -31,6 +31,11 @@
 if sklearn_check_version("1.1") and not sklearn_check_version("1.2"):
     from sklearn.utils import check_scalar
 
+if sklearn_check_version("1.6"):
+    from sklearn.utils.validation import validate_data
+else:
+    validate_data = sklearn_DBSCAN._validate_data
+
 
 class BaseDBSCAN(ABC):
     def _onedal_dbscan(self, **onedal_params):
@@ -85,7 +90,7 @@ def __init__(
 
     def _onedal_fit(self, X, y, sample_weight=None, queue=None):
         if sklearn_check_version("1.0"):
-            X = self._validate_data(X, force_all_finite=False)
+            X = validate_data(self, X, force_all_finite=False)
 
         onedal_params = {
             "eps": self.eps,

sklearnex/cluster/k_means.py

@@ -41,6 +41,11 @@
     from .._device_offload import dispatch, wrap_output_data
     from .._utils import PatchingConditionsChain
 
+    if sklearn_check_version("1.6"):
+        from sklearn.utils.validation import validate_data
+    else:
+        validate_data = sklearn_KMeans._validate_data
+
     @control_n_jobs(decorated_methods=["fit", "predict", "transform", "fit_transform"])
     class KMeans(sklearn_KMeans):
         __doc__ = sklearn_KMeans.__doc__
@@ -150,7 +155,8 @@ def fit(self, X, y=None, sample_weight=None):
             return self
 
         def _onedal_fit(self, X, _, sample_weight, queue=None):
-            X = self._validate_data(
+            X = validate_data(
+                self,
                 X,
                 accept_sparse="csr",
                 dtype=[np.float64, np.float32],
@@ -271,7 +277,8 @@ def predict(
         def _onedal_predict(self, X, sample_weight=None, queue=None):
             check_is_fitted(self)
 
-            X = self._validate_data(
+            X = validate_data(
+                self,
                 X,
                 accept_sparse="csr",
                 reset=False,
@@ -335,7 +342,8 @@ def score(self, X, y=None, sample_weight=None):
         def _onedal_score(self, X, y, sample_weight=None, queue=None):
             check_is_fitted(self)
 
-            X = self._validate_data(
+            X = validate_data(
+                self,
                 X,
                 accept_sparse="csr",
                 reset=False,

sklearnex/covariance/incremental_covariance.py

@@ -40,6 +40,11 @@
 if sklearn_check_version("1.2"):
     from sklearn.utils._param_validation import Interval
 
+if sklearn_check_version("1.6"):
+    from sklearn.utils.validation import validate_data
+else:
+    validate_data = BaseEstimator._validate_data
+
 
 @control_n_jobs(decorated_methods=["partial_fit", "fit", "_onedal_finalize_fit"])
 class IncrementalEmpiricalCovariance(BaseEstimator):
@@ -163,7 +168,8 @@ def _onedal_partial_fit(self, X, queue=None, check_input=True):
                 self._validate_params()
 
             if sklearn_check_version("1.0"):
-                X = self._validate_data(
+                X = validate_data(
+                    self,
                     X,
                     dtype=[np.float64, np.float32],
                     reset=first_pass,
@@ -204,7 +210,8 @@ def score(self, X_test, y=None):
 
         location = self.location_
         if sklearn_check_version("1.0"):
-            X = self._validate_data(
+            X = validate_data(
+                self,
                 X_test,
                 dtype=[np.float64, np.float32],
                 reset=False,
@@ -306,8 +313,12 @@ def _onedal_fit(self, X, queue=None):
 
         # finite check occurs on onedal side
         if sklearn_check_version("1.0"):
-            X = self._validate_data(
-                X, dtype=[np.float64, np.float32], copy=self.copy, force_all_finite=False
+            X = validate_data(
+                self,
+                X,
+                dtype=[np.float64, np.float32],
+                copy=self.copy,
+                force_all_finite=False,
             )
         else:
             X = check_array(

sklearnex/decomposition/pca.py

@@ -25,7 +25,7 @@
 
     import numpy as np
     from scipy.sparse import issparse
-    from sklearn.utils.validation import check_is_fitted
+    from sklearn.utils.validation import check_array, check_is_fitted
 
     from daal4py.sklearn._n_jobs_support import control_n_jobs
     from daal4py.sklearn._utils import sklearn_check_version
@@ -44,6 +44,11 @@
 
     from onedal.decomposition import PCA as onedal_PCA
 
+    if sklearn_check_version("1.6"):
+        from sklearn.utils.validation import validate_data
+    else:
+        validate_data = sklearn_PCA._validate_data
+
     @control_n_jobs(decorated_methods=["fit", "transform", "fit_transform"])
     class PCA(sklearn_PCA):
         __doc__ = sklearn_PCA.__doc__
@@ -133,7 +138,8 @@ def _fit(self, X):
             )
 
         def _onedal_fit(self, X, queue=None):
-            X = self._validate_data(
+            X = validate_data(
+                self,
                 X,
                 dtype=[np.float64, np.float32],
                 ensure_2d=True,
@@ -177,12 +183,17 @@ def transform(self, X):
         def _onedal_transform(self, X, queue=None):
             check_is_fitted(self)
             if sklearn_check_version("1.0"):
-                self._check_feature_names(X, reset=False)
-            X = self._validate_data(
-                X,
-                dtype=[np.float64, np.float32],
-                reset=False,
-            )
+                X = validate_data(
+                    self,
+                    X,
+                    dtype=[np.float64, np.float32],
+                    reset=False,
+                )
+            else:
+                X = check_array(
+                    X,
+                    dtype=[np.float64, np.float32],
+                )
             self._validate_n_features_in_after_fitting(X)
 
             return self._onedal_estimator.predict(X, queue=queue)

sklearnex/ensemble/_forest.py

@@ -20,7 +20,7 @@
 
 import numpy as np
 from scipy import sparse as sp
-from sklearn.base import clone
+from sklearn.base import BaseEstimator, clone
 from sklearn.ensemble import ExtraTreesClassifier as sklearn_ExtraTreesClassifier
 from sklearn.ensemble import ExtraTreesRegressor as sklearn_ExtraTreesRegressor
 from sklearn.ensemble import RandomForestClassifier as sklearn_RandomForestClassifier
@@ -67,12 +67,18 @@
 if sklearn_check_version("1.4"):
     from daal4py.sklearn.utils import _assert_all_finite
 
+if sklearn_check_version("1.6"):
+    from sklearn.utils.validation import validate_data
+else:
+    validate_data = BaseEstimator._validate_data
+
 
 class BaseForest(ABC):
     _onedal_factory = None
 
     def _onedal_fit(self, X, y, sample_weight=None, queue=None):
-        X, y = self._validate_data(
+        X, y = validate_data(
+            self,
             X,
             y,
             multi_output=True,
@@ -607,21 +613,7 @@ def predict_proba(self, X):
         # TODO:
         # _check_proba()
         # self._check_proba()
-        if sklearn_check_version("1.0"):
-            self._check_feature_names(X, reset=False)
-        if hasattr(self, "n_features_in_"):
-            try:
-                num_features = _num_features(X)
-            except TypeError:
-                num_features = _num_samples(X)
-            if num_features != self.n_features_in_:
-                raise ValueError(
-                    (
-                        f"X has {num_features} features, "
-                        f"but {self.__class__.__name__} is expecting "
-                        f"{self.n_features_in_} features as input"
-                    )
-                )
+
         return dispatch(
             self,
             "predict_proba",
@@ -797,7 +789,8 @@ def _onedal_predict(self, X, queue=None):
         check_is_fitted(self, "_onedal_estimator")
 
         if sklearn_check_version("1.0"):
-            X = self._validate_data(
+            X = validate_data(
+                self,
                 X,
                 dtype=[np.float64, np.float32],
                 force_all_finite=False,
@@ -810,6 +803,19 @@ def _onedal_predict(self, X, queue=None):
                 dtype=[np.float64, np.float32],
                 force_all_finite=False,
             )  # Warning, order of dtype matters
+            if hasattr(self, "n_features_in_"):
+                try:
+                    num_features = _num_features(X)
+                except TypeError:
+                    num_features = _num_samples(X)
+                if num_features != self.n_features_in_:
+                    raise ValueError(
+                        (
+                            f"X has {num_features} features, "
+                            f"but {self.__class__.__name__} is expecting "
+                            f"{self.n_features_in_} features as input"
+                        )
+                    )
             self._check_n_features(X, reset=False)
 
         res = self._onedal_estimator.predict(X, queue=queue)
@@ -819,7 +825,8 @@ def _onedal_predict_proba(self, X, queue=None):
         check_is_fitted(self, "_onedal_estimator")
 
         if sklearn_check_version("1.0"):
-            X = self._validate_data(
+            X = validate_data(
+                self,
                 X,
                 dtype=[np.float64, np.float32],
                 force_all_finite=False,
@@ -1123,7 +1130,8 @@ def _onedal_predict(self, X, queue=None):
         check_is_fitted(self, "_onedal_estimator")
 
         if sklearn_check_version("1.0"):
-            X = self._validate_data(
+            X = validate_data(
+                self,
                 X,
                 dtype=[np.float64, np.float32],
                 force_all_finite=False,

sklearnex/linear_model/incremental_linear.py

@@ -32,6 +32,11 @@
 if sklearn_check_version("1.2"):
     from sklearn.utils._param_validation import Interval
 
+if sklearn_check_version("1.6"):
+    from sklearn.utils.validation import validate_data
+else:
+    validate_data = BaseEstimator._validate_data
+
 from onedal.common.hyperparameters import get_hyperparameters
 
 from .._device_offload import dispatch, wrap_output_data
@@ -131,7 +136,8 @@ def _onedal_predict(self, X, queue=None):
             self._validate_params()
 
         if sklearn_check_version("1.0"):
-            X = self._validate_data(
+            X = validate_data(
+                self,
                 X,
                 dtype=[np.float64, np.float32],
                 copy=self.copy_X,
@@ -162,7 +168,8 @@ def _onedal_partial_fit(self, X, y, check_input=True, queue=None):
 
         if check_input:
             if sklearn_check_version("1.0"):
-                X, y = self._validate_data(
+                X, y = validate_data(
+                    self,
                     X,
                     y,
                     dtype=[np.float64, np.float32],
@@ -212,7 +219,8 @@ def _onedal_fit(self, X, y, queue=None):
             self._validate_params()
 
         if sklearn_check_version("1.0"):
-            X, y = self._validate_data(
+            X, y = validate_data(
+                self,
                 X,
                 y,
                 dtype=[np.float64, np.float32],