[enhancement] Add IncrementalEmpiricalCovariance to patch_map (#1674)

* initial hack * formatting * formatting * isorted * remove unnecessary lines * add another patch option * first corrections * Update covariance.py * Update covariance.py * Update dispatcher.py * Update incremental_covariance.py * Update covariance.py * Update covariance.py * Update covariance.py * Update incremental_covariance.py * Update test_patching.py * expand tests temporarily * reorient n_features_in_ * updates * added formatting * remove TODO * Update covariance.py * Update incremental_covariance.py * Update covariance.py * Update covariance.py * Update covariance.py * Update incremental_covariance.py * Update covariance.py * Update incremental_covariance.py * Update test_incremental_covariance.py * Update incremental_covariance.py * Update incremental_covariance.py * Update incremental_covariance.py * Update incremental_covariance.py * Update incremental_covariance.py * Update incremental_covariance.py * Update incremental_covariance.py * Update incremental_covariance.py * Update incremental_covariance.py * Update incremental_covariance.py * Update incremental_covariance.py * Update incremental_covariance.py * Update test_incremental_covariance.py * Update test_incremental_covariance.py * Update incremental_covariance.py * Update test_n_jobs_support.py * Update dispatcher.py * Update incremental_covariance.py * Update incremental_covariance.py * Update test_incremental_covariance.py * Update covariance.py * Update incremental_covariance.py * Update incremental_covariance.py * Update incremental_covariance.py * Update incremental_covariance.py * Update test_incremental_covariance.py * Update incremental_covariance.py * Update incremental_covariance.py * Update incremental_covariance.py * Update incremental_covariance.py * Update incremental_covariance.py * Update incremental_covariance.py * Update incremental_covariance.py * formatting * fix test_monkeypatch.py * switch order * formatting * generalize sklearn test patching in sklearnex * add wraps * add proper import * check_array -> _check_array * switch underscore * change text * add _check_array * linting * add warning * remove other PRs from this PR * forgotten None * final changes * and to or * add tests on recommendation * readd column_count * readd to patchmap * Update test_incremental_covariance.py * Update test_memory_usage.py * formatting * Update sklearnex/tests/test_memory_usage.py Co-authored-by: Alexander Andreev <alexander.andreev@intel.com> * change order to prevent fp64 issues * add assume_centered to daal calls * mistake * fix linting, change tests * allow for method=None * pre-protect for dpnp/dpctl * change daal_check_versions * change daal_check_versions * switch to zeros_like * formatting * fix c++ mistake * add testing for assume_centered * Update incremental_covariance.py * Update dispatcher.py * Update test_patching.py * add daal_check_version * Update incremental_covariance.py * Update incremental_covariance.py * Update covariance.py * Update incremental_covariance.py * Update incremental_covariance.py * Update incremental_covariance.py * Update incremental_covariance.py * Update incremental_covariance.py * Update covariance.py * Update test_incremental_covariance.py * Update incremental_covariance.py * Update test_patching.py * Update test_incremental_covariance.py * Update test_incremental_covariance.py * Update incremental_covariance.py * Update deselected_tests.yaml * formatting * fix bad merge * fix test failure * Update dispatcher.py * Update test_incremental_covariance.py * Update incremental_covariance.py * This may timeout CI * Update incremental_covariance.py * Update incremental_covariance.py * Update test_incremental_covariance.py * Update test_incremental_covariance.py * formatting * Update test_incremental_covariance.py * Update test_incremental_covariance.py * Update deselected_tests.yaml --------- Co-authored-by: Alexander Andreev <alexander.andreev@intel.com>
uxlfoundation · Apr 26, 2024 · 8c7a928 · 8c7a928
1 parent 57c3c73
commit 8c7a928
Show file tree

Hide file tree

Showing 6 changed files with 308 additions and 55 deletions.
diff --git a/deselected_tests.yaml b/deselected_tests.yaml
@@ -357,12 +357,13 @@ deselected_tests:
   # Need to rework getting policy to correctly obtain it for method without data (finalize_fit)
   # and avoid keeping it in class attribute, also need to investigate how to implement
   # partial result serialization
+  - tests/test_common.py::test_estimators[IncrementalEmpiricalCovariance()-check_estimators_pickle]
+  - tests/test_common.py::test_estimators[IncrementalEmpiricalCovariance()-check_estimators_pickle(readonly_memmap=True)]
   - tests/test_common.py::test_estimators[IncrementalLinearRegression()-check_estimators_pickle]
   - tests/test_common.py::test_estimators[IncrementalLinearRegression()-check_estimators_pickle(readonly_memmap=True)]
   # There are not enough data to run onedal backend
   - tests/test_common.py::test_estimators[IncrementalLinearRegression()-check_fit2d_1sample]
 
-
   # --------------------------------------------------------
   # No need to test daal4py patching
 reduced_tests:

diff --git a/onedal/covariance/incremental_covariance.py b/onedal/covariance/incremental_covariance.py
@@ -19,6 +19,7 @@
 from onedal import _backend
 
 from ..datatypes import _convert_to_supported, from_table, to_table
+from ..utils import _check_array
 from .covariance import BaseEmpiricalCovariance
 
 
@@ -37,6 +38,12 @@ class IncrementalEmpiricalCovariance(BaseEmpiricalCovariance):
         If True biased estimation of covariance is computed which equals to
         the unbiased one multiplied by (n_samples - 1) / n_samples.
 
+    assume_centered : bool, default=False
+        If True, data are not centered before computation.
+        Useful when working with data whose mean is almost, but not exactly
+        zero.
+        If False (default), data are centered before computation.
+
     Attributes
     ----------
     location_ : ndarray of shape (n_features,)
@@ -46,8 +53,11 @@ class IncrementalEmpiricalCovariance(BaseEmpiricalCovariance):
         Estimated covariance matrix
     """
 
-    def __init__(self, method="dense", bias=False):
-        super().__init__(method, bias)
+    def __init__(self, method="dense", bias=False, assume_centered=False):
+        super().__init__(method, bias, assume_centered)
+        self._reset()
+
+    def _reset(self):
         self._partial_result = self._get_backend(
             "covariance", None, "partial_compute_result"
         )
@@ -74,15 +84,16 @@ def partial_fit(self, X, y=None, queue=None):
         self : object
             Returns the instance itself.
         """
+        X = _check_array(X, dtype=[np.float64, np.float32], ensure_2d=True)
+
         if not hasattr(self, "_policy"):
             self._policy = self._get_policy(queue, X)
+
+        X = _convert_to_supported(self._policy, X)
+
         if not hasattr(self, "_dtype"):
             self._dtype = get_dtype(X)
-        X = make2d(X)
-        types = [np.float32, np.float64]
-        if get_dtype(X) not in types:
-            X = X.astype(np.float64)
-        X = _convert_to_supported(self._policy, X)
+
         params = self._get_onedal_params(self._dtype)
         table_X = to_table(X)
         self._partial_result = self._get_backend(