scikit-learn-intelex 2024.4.0__py39-none-manylinux1_x86_64.whl → 2024.6.0__py39-none-manylinux1_x86_64.whl

{scikit_learn_intelex-2024.4.0.dist-info → scikit_learn_intelex-2024.6.0.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: scikit-learn-intelex
-Version: 2024.4.0
+Version: 2024.6.0
 Summary: Intel(R) Extension for Scikit-learn is a seamless way to speed up your Scikit-learn application.
 Home-page: https://github.com/intel/scikit-learn-intelex
 Author: Intel Corporation
@@ -31,7 +31,7 @@ Classifier: Topic :: Software Development
 Requires-Python: >=3.7
 Description-Content-Type: text/markdown
 License-File: LICENSE.txt
-Requires-Dist: daal4py (==2024.4.0)
+Requires-Dist: daal4py (==2024.6.0)
 Requires-Dist: scikit-learn (>=0.22)
 
 

{scikit_learn_intelex-2024.4.0.dist-info → scikit_learn_intelex-2024.6.0.dist-info}/RECORD

@@ -1,38 +1,40 @@
 sklearnex/__init__.py,sha256=DVpGCMMQcUlrOGj_iy1jk3ZhCfa0_TFISYLqaUDRqC0,1780
 sklearnex/__main__.py,sha256=Le9BJq6aLEGSSoZwJLsOADxsV89ynBfA8BcoJHk9F24,1921
 sklearnex/_config.py,sha256=6WS3UuS4-0DxIJyGn7yQMosj-mGkLybLQrg3W7dED5o,3928
-sklearnex/_device_offload.py,sha256=EX9bRBV9tFxQBRf8mS9ntBEmB7tcIfql8sYnb9Zjlto,8639
+sklearnex/_device_offload.py,sha256=m8Hspffwx6Tn3f-OYLqwf5cUCKq4vZ3aSLmhY92qp08,8876
 sklearnex/_utils.py,sha256=EV4jC3plVdndsgrfPBsJZTzggrRdYWLwOpoIRWtTXt4,3812
-sklearnex/conftest.py,sha256=ZvcfQljZNBa3zlE1iITvuacHblugVtK6X80LwNXrnWI,2130
-sklearnex/dispatcher.py,sha256=Vpoy6kwOR0Y3ISfjpk0S1vEKhDZGLUKeMy-xwCBsCQs,14389
+sklearnex/conftest.py,sha256=ODuhlscC0HNGXiA8olEfHTDULzjevqG9_sn0yMGRkHg,2376
+sklearnex/dispatcher.py,sha256=XElvy9dDJ8XNOci8asKUnWXJpr6JROXHehdWBc_od3g,15876
 sklearnex/basic_statistics/__init__.py,sha256=-lNh54F2FO0BuHDBLjhW7zp3yBcgFZ-DrXHj2H5hLXE,920
 sklearnex/basic_statistics/basic_statistics.py,sha256=j5jBZ3DgXjoyjCBEFfYQHyl3LgLXEWWFnNHwInThdZw,796
 sklearnex/basic_statistics/incremental_basic_statistics.py,sha256=NNWF9zdxBz2jT7dzt5hL6HRvjQIGosvViQE3KJMF1L4,10048
-sklearnex/basic_statistics/tests/test_incremental_basic_statistics.py,sha256=GarYPjojsyhbOat5vOXF85AofyFa_VretX4Vntc01C8,14992
+sklearnex/basic_statistics/tests/test_incremental_basic_statistics.py,sha256=zTb_04DBGYSfwcpRaP0OJzi4Z6jz4jqV_kDRAp1x-no,14926
 sklearnex/cluster/__init__.py,sha256=TsDzdbKzEubaP86iY8a7mSW0ZQzgjzGYuPkhc5lZmlQ,853
-sklearnex/cluster/dbscan.py,sha256=JJaNVhV4dVK0xte9xzQpCAka80xVu3M7_SzmvD3k-EY,6736
+sklearnex/cluster/dbscan.py,sha256=EiAMRZq53KhGQ_d4_c867-frgG-pz8S1J88vd8hLAn4,6844
 sklearnex/cluster/k_means.py,sha256=1QKcFUQcnycu8kSD8uYSaDIuedHbxZsV_gvUfcEwVAM,806
-sklearnex/cluster/tests/test_dbscan.py,sha256=AgFoKwXFVyLKLvEtJVs0qlbBs1Ci3PcRft7f-6_ENOU,1464
+sklearnex/cluster/tests/test_dbscan.py,sha256=JYpwyuPkGQHdbE_IPbQv4qNj7clMm4UPdz_lrpRzKXE,1504
 sklearnex/cluster/tests/test_kmeans.py,sha256=H3a9NSHRRSASo0Eceo44Kjq6GwEoMqMzcubaNt1s1QQ,1213
 sklearnex/covariance/__init__.py,sha256=_c86wiKDGBH50ofN-_tn7lO4M2YVEWO5AHhJIfimUDk,859
-sklearnex/covariance/incremental_covariance.py,sha256=3s9gmLP48DPhlhnFcg2JB7RQ2sliSZdNroJkv8-1sIA,4526
-sklearnex/covariance/tests/test_incremental_covariance.py,sha256=JIa4p-1_RYvU2ZVLx27iz0OKp0Nwrl9sYCsT0Dlyi2I,5402
+sklearnex/covariance/incremental_covariance.py,sha256=0MvYlGnmuZ8_kRBaIkMzg3RyGdcSMjO-I7dhw2VpTsg,11010
+sklearnex/covariance/tests/test_incremental_covariance.py,sha256=ZbEfup4ICm278RW4hZIHPciOiqhFhx_k1l3lpnw0M6s,6763
 sklearnex/decomposition/__init__.py,sha256=RJBYDWyvnn5DA-j5r4IqVdXLp6H4mdPySsSjnXhlv-U,805
-sklearnex/decomposition/pca.py,sha256=ffR22wH2n5R8SYflEmOPDPc-QpTomiAMOegV0IMbslg,12654
-sklearnex/decomposition/tests/test_pca.py,sha256=FJYMosxpQpdOo71QD0jVx47oWEhvVNdut_5EOxTyT_0,2201
+sklearnex/decomposition/pca.py,sha256=JkyxsSz_8vVGUjJrfxyAwwY1Yf2uht-qxwOisEHv9mY,15550
+sklearnex/decomposition/tests/test_pca.py,sha256=EoCgpSojE2S2e7hOUwW0Bh3vVGTUywawAhU7ThVAlW0,2319
 sklearnex/doc/third-party-programs.txt,sha256=qA1XbOqYkMRnp29p8VxXjfcH0kHE8NSO5s2heea83-8,21773
 sklearnex/ensemble/__init__.py,sha256=2xB2KfG7l7uJv0p3kfVhrs4KLM174SCsigZhYqwSYAA,1035
-sklearnex/ensemble/_forest.py,sha256=OHxqhYhkdjzhhEeWLlE0_IXBrWmKCUJoP_B1gQcmTQM,70002
-sklearnex/ensemble/tests/test_forest.py,sha256=r1GSPQuVVuWYoNB3ZLiGKfMSwq3lp5i6k-Rgr7InonQ,4456
+sklearnex/ensemble/_forest.py,sha256=tElkh0jezrp9QK2T3_kDCgye6El9W6hOlJepaeMIOM4,73494
+sklearnex/ensemble/tests/test_forest.py,sha256=UXj6pfSuOr8wm_KZciyE9zt6hh-we1U8s0XU9bShqHI,4735
 sklearnex/glob/__main__.py,sha256=--FAjkhh5E6a1TFOGu6_BM3MHe78jP8oSpl0xKiRtTI,2531
-sklearnex/glob/dispatcher.py,sha256=IEEPhAOCVzC2JDFvYtijbiPAbaUY5RrCAhFLRjMMe1w,3018
-sklearnex/linear_model/__init__.py,sha256=5XZDZh8R0SmT8D8ZtSjW7-MKRO7l1jOZ8CUnt_OzHe4,1047
+sklearnex/glob/dispatcher.py,sha256=o6XKGKM3M91F7FlXBOt1IhnpWQK4R1VY2WS-0uIghcw,3906
+sklearnex/linear_model/__init__.py,sha256=mzZ0EaBhDH66ETNt2vylznSoZbCYexgL2qE_jKppYYc,1144
 sklearnex/linear_model/coordinate_descent.py,sha256=uZOHIKfFlHoMlNXZCh2MAnZE30fRZlmtcF7UsZQ3Vq4,822
-sklearnex/linear_model/linear.py,sha256=0A2kCTpsGZK8IVkbIhYoAXqMAxHOyY7dO5JLxbNWmLw,11159
+sklearnex/linear_model/incremental_linear.py,sha256=iDad4Z4xMDImvG-Jgtt0Hz3Qp9MVhSi-IjUPcV0Y_ks,16101
+sklearnex/linear_model/linear.py,sha256=SUoFFvpl4Eb308sYBp6FNbZls-4G69eg8wSeuPs0oAg,11765
 sklearnex/linear_model/logistic_path.py,sha256=Nq3JPXSzE4bjamnP3gGQtsMKks2v7s6bSiuYRnFqrRw,849
-sklearnex/linear_model/logistic_regression.py,sha256=6TRaDJ8nIHxjofIfUvmeJOP9_EwmUgVKKK1sxe9vS5A,14171
+sklearnex/linear_model/logistic_regression.py,sha256=jmsNs7MYHm-PncuG20m_TSqXbM5jXx_vHsoH-hReEFQ,14271
 sklearnex/linear_model/ridge.py,sha256=0oxlM5McYYvl0KxK9OIGJKM6lOADuFSPTdfx-efJNTI,810
-sklearnex/linear_model/tests/test_linear.py,sha256=kLoOcP9p0iCK2PhU438tGuPGuuBekF37FkA-YBcBMpM,4319
+sklearnex/linear_model/tests/test_incremental_linear.py,sha256=e0ZADjB0myq1QcdwYxlVYl6tGFs4tVZIfBoV1xkdFuw,7337
+sklearnex/linear_model/tests/test_linear.py,sha256=9YQgyYmcGSL2rEgIfhZZxTIlj5v-Z6-ygBqYR6ly3oE,4357
 sklearnex/linear_model/tests/test_logreg.py,sha256=mUYDwTDUekERlzxjnVeOeCUcNv4KJAXdTIiELZNSaDc,3283
 sklearnex/manifold/__init__.py,sha256=3FU5fY1YvHAyNeSlUHsZfPnAbvlAUtZpum0Obmok2KE,809
 sklearnex/manifold/t_sne.py,sha256=U6F_dGB6taVuEJvTnVBWD-sri3x3m0Khu3HI4wXidhg,805
@@ -48,16 +50,19 @@ sklearnex/neighbors/__init__.py,sha256=_GMJ2L-6YechRYI2fOFwHjy0kebUncE81z4FmdXUl
 sklearnex/neighbors/_lof.py,sha256=5MPkWIb6FS55Q8xWzgc22Ec_PsouuN94SPovt-vsBGE,8648
 sklearnex/neighbors/common.py,sha256=8DxDoYtXtEM4RNoMCimpTVSDOOxUIJlNVOQvBXBhkd4,10875
 sklearnex/neighbors/knn_classification.py,sha256=ONoBbtwb5xoVLcpg5COpRNu1ZBOQ2EDh03RzbWDy5yo,8537
-sklearnex/neighbors/knn_regression.py,sha256=YBjvh6h56OaJHSZNGq8ZV8lsB-XpSWduEcVeHRuPorQ,6692
+sklearnex/neighbors/knn_regression.py,sha256=ge8VAMa8aJ58M5ccN5NMILFHuRYtYKzyoLF8fxxgGfo,7462
 sklearnex/neighbors/knn_unsupervised.py,sha256=qkj-hSEq8QetYsWlofnik3PpFvo1iBUbIOhywo8wFWk,5362
-sklearnex/neighbors/tests/test_neighbors.py,sha256=gcJeJwDoT7wvyvk423k9TDGTGfbXUSTQETUT-VYQ74U,3409
-sklearnex/preview/__init__.py,sha256=OXC_k2kudA13rVesXb5ZlEC7_mKUS7uUra_BR-Tin30,780
+sklearnex/neighbors/tests/test_neighbors.py,sha256=yAlMnLt9GrdT6Ceph5B7iFuMJXpDURiHWTE99oO8EDw,3417
+sklearnex/preview/__init__.py,sha256=1QcbV6xCSP7QCXRxYLVPc_b4lxE0cQiyTrMm4xOnosM,797
 sklearnex/preview/cluster/__init__.py,sha256=FONzOuTGAb5NwdfFhLco99P_VccRk9NBkDHU3EKuAIs,794
 sklearnex/preview/cluster/_common.py,sha256=bgpzijxgyrAtdNCCHyCNsZ-N70KQYNwMuoCTNdUna6o,2718
 sklearnex/preview/cluster/k_means.py,sha256=RdF9mwgvZ1xgyeLhCQA2WaPvPsS1ndktXXcaNj6S460,10464
 sklearnex/preview/covariance/__init__.py,sha256=DPYTk8lwcVphtwU8J3CyUYH8Uz6Zr0Uz4S6nn-RY5iM,825
 sklearnex/preview/covariance/covariance.py,sha256=NjRTWLie7UbpfN16wGSxJSRHRhUnvTn03cPPc8PpPd8,5131
 sklearnex/preview/covariance/tests/test_covariance.py,sha256=FOgvhJxFQo9sNG4Uo-ebuw3ZXJ9tSxdk8qKm0ZJzTdc,2487
+sklearnex/preview/decomposition/__init__.py,sha256=9VcJPWKgSrWDFEXUY6ZCpAT2XGbOVA4a1j_XgfJBnTM,839
+sklearnex/preview/decomposition/incremental_pca.py,sha256=SsY1-AQt0mFTJGP5yzVxZvopNz2tSeXUO9p9c_3uVus,7820
+sklearnex/preview/decomposition/tests/test_incremental_pca.py,sha256=tsAlM18nzfIQxic7Ry986Ue0ovUdbopWFNckqQLK5xU,10776
 sklearnex/spmd/__init__.py,sha256=ChQy2kEWlo4KGvs0RnbPoPEVhdgl8URV099B1rZtF5Y,889
 sklearnex/spmd/basic_statistics/__init__.py,sha256=NA5RGlwcp27UEeCgz0ngzYYd3MAIRpxlTy86uhM2-RE,821
 sklearnex/spmd/basic_statistics/basic_statistics.py,sha256=_dQ9mhVYxeuChATEpAmHpXDpgW3YvtK1qrG-kLr2MtI,886
@@ -76,26 +81,28 @@ sklearnex/spmd/linear_model/logistic_regression.py,sha256=q_HkfWcg0RgFbk2t9FeV0Z
 sklearnex/spmd/neighbors/__init__.py,sha256=S16sH8N_18LN_8AgC_wGK5EDSNyuN-F-gI6GVlaiWVE,906
 sklearnex/spmd/neighbors/neighbors.py,sha256=SiKAS_RVt34MUcGytBS5pHnI_5vFNxJn8jqt1MOhDh8,940
 sklearnex/svm/__init__.py,sha256=f3e4ZFwZfx6MsXsn94VK1xVm6mWKT5XCiHczo6zNyAQ,1057
-sklearnex/svm/_common.py,sha256=52csEWTYWEVgUVWUm3RjX1VD_5VyyawXBvc7lBBp5qY,7010
-sklearnex/svm/nusvc.py,sha256=J4F_Tm9oYC-WK1Nf_CNqSnvuYEEZZ9xa3VY-x0G_2cw,9871
-sklearnex/svm/nusvr.py,sha256=ZFBnQzMgD4aRSk3BRIxwKfwErVyDJnNoR7vrIMm5WLk,3655
-sklearnex/svm/svc.py,sha256=u9wQ8dsvQvO4XP_3JSfYFBNJcuDyv3HVG5NY6jZqhZk,11125
-sklearnex/svm/svr.py,sha256=ldNWehrU4vgqX_0T8splmBjPsQln7w_h8L4XNpIgP6A,3655
+sklearnex/svm/_common.py,sha256=wH4LT6QIoe7xM3btaUBC7fpKJcqKLfNHJj16rAaEZ1k,12797
+sklearnex/svm/nusvc.py,sha256=-SfYz2bUMLaosYg_qeTBYf1Ra-bFuGmI62xIerf3XeE,11262
+sklearnex/svm/nusvr.py,sha256=9nbRaPCoDQ5SBGH_jLgcaX_PWGDh667w-gW5X2YbHBM,4764
+sklearnex/svm/svc.py,sha256=fZ9oVpKkgH5OYBXo4mPdZ-8cP07hSzPHoJsyItpXmcQ,12470
+sklearnex/svm/svr.py,sha256=NIWVUu7NjV4Fe4L-Y4vIqd3KePFdMJ1-jupeH4ZvNsg,4733
 sklearnex/svm/tests/test_svm.py,sha256=Ru-aGNGCsRJts7SEEYbnKcVqUx-DqPyUtw-hEoMVpW8,4190
-sklearnex/tests/_utils.py,sha256=hc3LxZDhlncONCGxqvWajQE7iVnSnt98Q9bem-KJDGI,5386
+sklearnex/tests/_utils.py,sha256=zRNpggSrZs4H0L-__UuiYRPm_ASqrqWm226ZQzlBQ7I,6252
+sklearnex/tests/test_common.py,sha256=MYx0y7oomArxWu9qe2zNs7YL-ScPGFnzvw8PZ1or04A,1813
 sklearnex/tests/test_config.py,sha256=SnSJjxAAysISDyC3bYKSJiRHStkB9X-yjLeF11LpRog,1372
-sklearnex/tests/test_memory_usage.py,sha256=LwIuUgG7CUZ2kOk1XlxyaK7O3OlyoVwx40TgljaYs5Q,7327
+sklearnex/tests/test_memory_usage.py,sha256=FTtQgaa7p5ScDGscB93LkpU4B3DzfqnVMi84b2vSL30,10957
 sklearnex/tests/test_monkeypatch.py,sha256=_9s_4jFvbttOf8uCuBrS4rn_AzQdFlKyRlthnGacUcU,9669
-sklearnex/tests/test_n_jobs_support.py,sha256=fNGlQz882qAJoJ56ymbID0hpIsO_7PHLmXxSew2JA_k,4124
+sklearnex/tests/test_n_jobs_support.py,sha256=NziTP4GwZEDoBe1CDvhHZnp8JpwjYQmCNvXEDfS7Wo4,4313
 sklearnex/tests/test_parallel.py,sha256=0zzlh2VJZWcHw5W4QSUthtAygOb6K0Bg1ekNsrdGJQE,1770
-sklearnex/tests/test_patching.py,sha256=Gc4MChHY6lVUFkm3sgPF4RDd1uOqHQFfh9ac60d4vfQ,14299
-sklearnex/tests/test_run_to_run_stability_tests.py,sha256=4HDOeJruA1EDILbyQJtsHFmEXC0D1upSHuOT-KyTlEc,14008
+sklearnex/tests/test_patching.py,sha256=LKi-x0ELu_y5HEa86UYhDzOalJphiEBtEe5own89PEs,14782
+sklearnex/tests/test_run_to_run_stability.py,sha256=49WnqTcAaGGa5eP7RBwEePvr_dA9hfYsCK4-pZA8OPw,9984
 sklearnex/utils/__init__.py,sha256=FFlM8zc2qnSdPYpLjSt4Ma_9FjC8qk8wXJa-5B8hCs0,898
-sklearnex/utils/_namespace.py,sha256=lWzvzcObxq350A-Ms8JyTbaQStR8rgH4DgQYhWruHL4,3166
+sklearnex/utils/_namespace.py,sha256=ohr8gOgEFgrccLecllMVYQPqbqqyye9uT-cWLtyxHFs,3167
 sklearnex/utils/parallel.py,sha256=VBcS-KUdyq7XpJUN6ygmNjyWtYLroghbvCxQ8nVU3YI,2085
 sklearnex/utils/validation.py,sha256=fjfhQiKnBQnD7LCBlacMyvsrhGnlMLRXk5Q69uoZIP4,827
-scikit_learn_intelex-2024.4.0.dist-info/LICENSE.txt,sha256=7micbUpzQXphq9e_2oL7PpZcvoXzPuQHIDEXyKXC81s,10797
-scikit_learn_intelex-2024.4.0.dist-info/METADATA,sha256=We3kZekomEyuXZpbrpq05hqphdjy8S5Q_geoj0Mb93c,12449
-scikit_learn_intelex-2024.4.0.dist-info/WHEEL,sha256=rxMEw7jRW2YnjujGudhK0a-ZC6J_VMIRJJ7uhbmewD4,107
-scikit_learn_intelex-2024.4.0.dist-info/top_level.txt,sha256=kzKChSWGJEYFmdj5PwE63HNuP_PVOhWfD32ytH9rL9Q,10
-scikit_learn_intelex-2024.4.0.dist-info/RECORD,,
+sklearnex/utils/tests/test_finite.py,sha256=AzJRY71X0VvDUicUI8Ey9Le6_yKp5O-3ZikhDVJNWms,2943
+scikit_learn_intelex-2024.6.0.dist-info/LICENSE.txt,sha256=7micbUpzQXphq9e_2oL7PpZcvoXzPuQHIDEXyKXC81s,10797
+scikit_learn_intelex-2024.6.0.dist-info/METADATA,sha256=HAyHGB0wdXHb5FaRHSiitrRZgs5yDtXkjfAigVtQJxA,12449
+scikit_learn_intelex-2024.6.0.dist-info/WHEEL,sha256=rxMEw7jRW2YnjujGudhK0a-ZC6J_VMIRJJ7uhbmewD4,107
+scikit_learn_intelex-2024.6.0.dist-info/top_level.txt,sha256=kzKChSWGJEYFmdj5PwE63HNuP_PVOhWfD32ytH9rL9Q,10
+scikit_learn_intelex-2024.6.0.dist-info/RECORD,,

sklearnex/_device_offload.py

@@ -127,8 +127,15 @@ def _transfer_to_host(queue, *data):
                 queue = usm_iface["syclobj"]
 
             buffer = as_usm_memory(item).copy_to_host()
+            order = "C"
+            if usm_iface["strides"] is not None:
+                if usm_iface["strides"][0] < usm_iface["strides"][1]:
+                    order = "F"
             item = np.ndarray(
-                shape=usm_iface["shape"], dtype=usm_iface["typestr"], buffer=buffer
+                shape=usm_iface["shape"],
+                dtype=usm_iface["typestr"],
+                buffer=buffer,
+                order=order,
             )
             has_usm_data = True
         else:

sklearnex/basic_statistics/tests/test_incremental_basic_statistics.py

@@ -165,7 +165,7 @@ def test_partial_fit_multiple_options_on_random_data(
             expected_sum(X),
         )
 
-    tol = 1e-5 if res_mean.dtype == np.float32 else 1e-7
+    tol = 3e-4 if res_mean.dtype == np.float32 else 1e-7
     assert_allclose(gtr_mean, res_mean, atol=tol)
     assert_allclose(gtr_max, res_max, atol=tol)
     assert_allclose(gtr_sum, res_sum, atol=tol)
@@ -208,7 +208,6 @@ def test_partial_fit_all_option_on_random_data(
 
     for option in options_and_tests:
         result_option, function, tols = option
-        print(result_option)
         fp32tol, fp64tol = tols
         res = getattr(result, result_option)
         if weighted:
@@ -301,7 +300,7 @@ def test_fit_single_option_on_random_data(
 @pytest.mark.parametrize("column_count", [10, 100])
 @pytest.mark.parametrize("weighted", [True, False])
 @pytest.mark.parametrize("dtype", [np.float32, np.float64])
-def test_partial_fit_multiple_options_on_random_data(
+def test_fit_multiple_options_on_random_data(
     dataframe, queue, num_batches, row_count, column_count, weighted, dtype
 ):
     seed = 77
@@ -375,7 +374,6 @@ def test_fit_all_option_on_random_data(
 
     for option in options_and_tests:
         result_option, function, tols = option
-        print(result_option)
         fp32tol, fp64tol = tols
         res = getattr(result, result_option)
         if weighted:

sklearnex/cluster/dbscan.py

@@ -85,6 +85,9 @@ class DBSCAN(sklearn_DBSCAN, BaseDBSCAN):
         self.n_jobs = n_jobs
 
     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)
+
         onedal_params = {
             "eps": self.eps,
             "min_samples": self.min_samples,

sklearnex/cluster/tests/test_dbscan.py

@@ -18,16 +18,18 @@ import numpy as np
 import pytest
 from numpy.testing import assert_allclose
 
+from onedal.tests.utils._dataframes_support import (
+    _convert_to_dataframe,
+    get_dataframes_and_queues,
+)
 
-# TODO:
-# adding this parameterized testing
-# somehow breaks other test with preview module patch:
-# sklearnex/tests/test_monkeypatch.py::test_preview_namespace.
-# @pytest.mark.parametrize("dataframe,queue", get_dataframes_and_queues())
-def test_sklearnex_import_dbscan():
+
+@pytest.mark.parametrize("dataframe,queue", get_dataframes_and_queues())
+def test_sklearnex_import_dbscan(dataframe, queue):
     from sklearnex.cluster import DBSCAN
 
     X = np.array([[1, 2], [2, 2], [2, 3], [8, 7], [8, 8], [25, 80]])
+    X = _convert_to_dataframe(X, sycl_queue=queue, target_df=dataframe)
     dbscan = DBSCAN(eps=3, min_samples=2).fit(X)
     assert "sklearnex" in dbscan.__module__
 

sklearnex/conftest.py

@@ -19,7 +19,8 @@ import logging
 
 import pytest
 
-from sklearnex import patch_sklearn, unpatch_sklearn
+from daal4py.sklearn._utils import sklearn_check_version
+from sklearnex import config_context, patch_sklearn, unpatch_sklearn
 
 
 def pytest_configure(config):
@@ -61,3 +62,12 @@ def with_sklearnex():
     patch_sklearn()
     yield
     unpatch_sklearn()
+
+
+@pytest.fixture
+def with_array_api():
+    if sklearn_check_version("1.2"):
+        with config_context(array_api_dispatch=True):
+            yield
+    else:
+        yield

sklearnex/covariance/incremental_covariance.py

@@ -14,18 +14,33 @@
 # limitations under the License.
 # ===============================================================================
 
+import numbers
+import warnings
+
 import numpy as np
+from scipy import linalg
+from sklearn.base import BaseEstimator
+from sklearn.covariance import EmpiricalCovariance as sklearn_EmpiricalCovariance
 from sklearn.utils import check_array, gen_batches
 
 from daal4py.sklearn._n_jobs_support import control_n_jobs
+from daal4py.sklearn._utils import daal_check_version, sklearn_check_version
 from onedal._device_offload import support_usm_ndarray
 from onedal.covariance import (
     IncrementalEmpiricalCovariance as onedal_IncrementalEmpiricalCovariance,
 )
+from sklearnex import config_context
+
+from .._device_offload import dispatch, wrap_output_data
+from .._utils import PatchingConditionsChain, register_hyperparameters
+from ..metrics import pairwise_distances
 
+if sklearn_check_version("1.2"):
+    from sklearn.utils._param_validation import Interval
 
-@control_n_jobs(decorated_methods=["partial_fit"])
-class IncrementalEmpiricalCovariance:
+
+@control_n_jobs(decorated_methods=["partial_fit", "fit", "_onedal_finalize_fit"])
+class IncrementalEmpiricalCovariance(BaseEstimator):
     """
     Incremental estimator for covariance.
     Allows to compute empirical covariance estimated by maximum
@@ -33,12 +48,25 @@ class IncrementalEmpiricalCovariance:
 
     Parameters
     ----------
+    store_precision : bool, default=False
+        Specifies if the estimated precision is stored.
+
+    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.
+
     batch_size : int, default=None
         The number of samples to use for each batch. Only used when calling
         ``fit``. If ``batch_size`` is ``None``, then ``batch_size``
         is inferred from the data and set to ``5 * n_features``, to provide a
         balance between approximation accuracy and memory consumption.
 
+    copy : bool, default=True
+        If False, X will be overwritten. ``copy=False`` can be used to
+        save memory but is unsafe for general use.
+
     Attributes
     ----------
     location_ : ndarray of shape (n_features,)
@@ -46,44 +74,130 @@ class IncrementalEmpiricalCovariance:
 
     covariance_ : ndarray of shape (n_features, n_features)
         Estimated covariance matrix
+
+    n_samples_seen_ : int
+        The number of samples processed by the estimator. Will be reset on
+        new calls to fit, but increments across ``partial_fit`` calls.
+
+    batch_size_ : int
+        Inferred batch size from ``batch_size``.
+
+    n_features_in_ : int
+        Number of features seen during :term:`fit` `partial_fit`.
     """
 
     _onedal_incremental_covariance = staticmethod(onedal_IncrementalEmpiricalCovariance)
 
-    def __init__(self, batch_size=None):
-        self._need_to_finalize = False  # If True then finalize compute should
-        #      be called to obtain covariance_ or location_ from partial compute data
+    if sklearn_check_version("1.2"):
+        _parameter_constraints: dict = {
+            "store_precision": ["boolean"],
+            "assume_centered": ["boolean"],
+            "batch_size": [Interval(numbers.Integral, 1, None, closed="left"), None],
+            "copy": ["boolean"],
+        }
+
+    get_precision = sklearn_EmpiricalCovariance.get_precision
+    error_norm = wrap_output_data(sklearn_EmpiricalCovariance.error_norm)
+    score = wrap_output_data(sklearn_EmpiricalCovariance.score)
+
+    def __init__(
+        self, *, store_precision=False, assume_centered=False, batch_size=None, copy=True
+    ):
+        self.assume_centered = assume_centered
+        self.store_precision = store_precision
         self.batch_size = batch_size
+        self.copy = copy
+
+    def _onedal_supported(self, method_name, *data):
+        patching_status = PatchingConditionsChain(
+            f"sklearn.covariance.{self.__class__.__name__}.{method_name}"
+        )
+        return patching_status
 
     def _onedal_finalize_fit(self):
         assert hasattr(self, "_onedal_estimator")
         self._onedal_estimator.finalize_fit()
         self._need_to_finalize = False
 
-    def _onedal_partial_fit(self, X, queue):
+        if not daal_check_version((2024, "P", 400)) and self.assume_centered:
+            location = self._onedal_estimator.location_[None, :]
+            self._onedal_estimator.covariance_ += np.dot(location.T, location)
+            self._onedal_estimator.location_ = np.zeros_like(np.squeeze(location))
+        if self.store_precision:
+            self.precision_ = linalg.pinvh(
+                self._onedal_estimator.covariance_, check_finite=False
+            )
+        else:
+            self.precision_ = None
+
+    @property
+    def covariance_(self):
+        if hasattr(self, "_onedal_estimator"):
+            if self._need_to_finalize:
+                self._onedal_finalize_fit()
+            return self._onedal_estimator.covariance_
+        else:
+            raise AttributeError(
+                f"'{self.__class__.__name__}' object has no attribute 'covariance_'"
+            )
+
+    @property
+    def location_(self):
+        if hasattr(self, "_onedal_estimator"):
+            if self._need_to_finalize:
+                self._onedal_finalize_fit()
+            return self._onedal_estimator.location_
+        else:
+            raise AttributeError(
+                f"'{self.__class__.__name__}' object has no attribute 'location_'"
+            )
+
+    def _onedal_partial_fit(self, X, queue=None, check_input=True):
+
+        first_pass = not hasattr(self, "n_samples_seen_") or self.n_samples_seen_ == 0
+
+        # finite check occurs on onedal side
+        if check_input:
+            if sklearn_check_version("1.2"):
+                self._validate_params()
+
+            if sklearn_check_version("1.0"):
+                X = self._validate_data(
+                    X,
+                    dtype=[np.float64, np.float32],
+                    reset=first_pass,
+                    copy=self.copy,
+                    force_all_finite=False,
+                )
+            else:
+                X = check_array(
+                    X,
+                    dtype=[np.float64, np.float32],
+                    copy=self.copy,
+                    force_all_finite=False,
+                )
+
         onedal_params = {
             "method": "dense",
             "bias": True,
+            "assume_centered": self.assume_centered,
         }
         if not hasattr(self, "_onedal_estimator"):
             self._onedal_estimator = self._onedal_incremental_covariance(**onedal_params)
-        self._onedal_estimator.partial_fit(X, queue)
-        self._need_to_finalize = True
+        try:
+            if first_pass:
+                self.n_samples_seen_ = X.shape[0]
+                self.n_features_in_ = X.shape[1]
+            else:
+                self.n_samples_seen_ += X.shape[0]
 
-    @property
-    def covariance_(self):
-        if self._need_to_finalize:
-            self._onedal_finalize_fit()
-        return self._onedal_estimator.covariance_
+            self._onedal_estimator.partial_fit(X, queue)
+        finally:
+            self._need_to_finalize = True
 
-    @property
-    def location_(self):
-        if self._need_to_finalize:
-            self._onedal_finalize_fit()
-        return self._onedal_estimator.location_
+        return self
 
-    @support_usm_ndarray()
-    def partial_fit(self, X, queue=None):
+    def partial_fit(self, X, y=None, check_input=True):
         """
         Incremental fit with X. All of X is processed as a single batch.
 
@@ -93,16 +207,29 @@ class IncrementalEmpiricalCovariance:
             Training data, where `n_samples` is the number of samples and
             `n_features` is the number of features.
 
+        y : Ignored
+            Not used, present for API consistency by convention.
+
+        check_input : bool, default=True
+            Run check_array on X.
+
         Returns
         -------
         self : object
             Returns the instance itself.
         """
-        X = check_array(X, dtype=[np.float64, np.float32])
-        self._onedal_partial_fit(X, queue)
-        return self
+        return dispatch(
+            self,
+            "partial_fit",
+            {
+                "onedal": self.__class__._onedal_partial_fit,
+                "sklearn": None,
+            },
+            X,
+            check_input=check_input,
+        )
 
-    def fit(self, X, queue=None):
+    def fit(self, X, y=None):
         """
         Fit the model with X, using minibatches of size batch_size.
 
@@ -112,19 +239,79 @@ class IncrementalEmpiricalCovariance:
             Training data, where `n_samples` is the number of samples and
             `n_features` is the number of features.
 
+        y : Ignored
+            Not used, present for API consistency by convention.
+
         Returns
         -------
         self : object
             Returns the instance itself.
         """
-        n_samples, n_features = X.shape
-        if self.batch_size is None:
-            batch_size_ = 5 * n_features
+
+        return dispatch(
+            self,
+            "fit",
+            {
+                "onedal": self.__class__._onedal_fit,
+                "sklearn": None,
+            },
+            X,
+        )
+
+    def _onedal_fit(self, X, queue=None):
+        self.n_samples_seen_ = 0
+        if hasattr(self, "_onedal_estimator"):
+            self._onedal_estimator._reset()
+
+        if sklearn_check_version("1.2"):
+            self._validate_params()
+
+        # 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
+            )
         else:
-            batch_size_ = self.batch_size
-        for batch in gen_batches(n_samples, batch_size_):
+            X = check_array(
+                X, dtype=[np.float64, np.float32], copy=self.copy, force_all_finite=False
+            )
+            self.n_features_in_ = X.shape[1]
+
+        self.batch_size_ = self.batch_size if self.batch_size else 5 * self.n_features_in_
+
+        if X.shape[0] == 1:
+            warnings.warn(
+                "Only one sample available. You may want to reshape your data array"
+            )
+
+        for batch in gen_batches(X.shape[0], self.batch_size_):
             X_batch = X[batch]
-            self.partial_fit(X_batch, queue=queue)
+            self._onedal_partial_fit(X_batch, queue=queue, check_input=False)
 
         self._onedal_finalize_fit()
+
         return self
+
+    # expose sklearnex pairwise_distances if mahalanobis distance eventually supported
+    @wrap_output_data
+    def mahalanobis(self, X):
+        if sklearn_check_version("1.0"):
+            self._validate_data(X, reset=False, copy=self.copy)
+        else:
+            check_array(X, copy=self.copy)
+
+        precision = self.get_precision()
+        with config_context(assume_finite=True):
+            # compute mahalanobis distances
+            dist = pairwise_distances(
+                X, self.location_[np.newaxis, :], metric="mahalanobis", VI=precision
+            )
+
+        return np.reshape(dist, (len(X),)) ** 2
+
+    _onedal_cpu_supported = _onedal_supported
+    _onedal_gpu_supported = _onedal_supported
+
+    mahalanobis.__doc__ = sklearn_EmpiricalCovariance.mahalanobis.__doc__
+    error_norm.__doc__ = sklearn_EmpiricalCovariance.error_norm.__doc__
+    score.__doc__ = sklearn_EmpiricalCovariance.score.__doc__