scikit-learn-intelex 2024.3.0__py310-none-manylinux1_x86_64.whl → 2024.5.0__py310-none-manylinux1_x86_64.whl

{scikit_learn_intelex-2024.3.0.dist-info → scikit_learn_intelex-2024.5.0.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: scikit-learn-intelex
-Version: 2024.3.0
+Version: 2024.5.0
 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 (==2024.3.0)
+Requires-Dist: daal4py (==2024.5.0)
 Requires-Dist: scikit-learn (>=0.22)
 
 

{scikit_learn_intelex-2024.3.0.dist-info → scikit_learn_intelex-2024.5.0.dist-info}/RECORD

@@ -1,36 +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=J0tZqj6tfvIFonWR01PadtTLgloQk_QEfXeCoqEvJlk,7710
+sklearnex/_device_offload.py,sha256=m8Hspffwx6Tn3f-OYLqwf5cUCKq4vZ3aSLmhY92qp08,8876
 sklearnex/_utils.py,sha256=EV4jC3plVdndsgrfPBsJZTzggrRdYWLwOpoIRWtTXt4,3812
 sklearnex/conftest.py,sha256=ZvcfQljZNBa3zlE1iITvuacHblugVtK6X80LwNXrnWI,2130
-sklearnex/dispatcher.py,sha256=9NPax55jPByCK0zzIvvE4wtmEJ7pCMrDh7IPeP2-vZ8,14389
-sklearnex/basic_statistics/__init__.py,sha256=NA5RGlwcp27UEeCgz0ngzYYd3MAIRpxlTy86uhM2-RE,821
+sklearnex/dispatcher.py,sha256=FtPF7nezu67wKhG33FAxJCWK8jCj65G87k7AITOIxtw,15390
+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=zTb_04DBGYSfwcpRaP0OJzi4Z6jz4jqV_kDRAp1x-no,14926
 sklearnex/cluster/__init__.py,sha256=TsDzdbKzEubaP86iY8a7mSW0ZQzgjzGYuPkhc5lZmlQ,853
 sklearnex/cluster/dbscan.py,sha256=JJaNVhV4dVK0xte9xzQpCAka80xVu3M7_SzmvD3k-EY,6736
 sklearnex/cluster/k_means.py,sha256=1QKcFUQcnycu8kSD8uYSaDIuedHbxZsV_gvUfcEwVAM,806
 sklearnex/cluster/tests/test_dbscan.py,sha256=AgFoKwXFVyLKLvEtJVs0qlbBs1Ci3PcRft7f-6_ENOU,1464
 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=XLUUM-bGeXiFXc81uUxNEIYwQ6e5edmjQ94wqnLt_xg,12829
-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=-3r3QwALcZregRq-MX8aUFSVaHu0D2ffGt-V5JOnNgE,2223
 sklearnex/doc/third-party-programs.txt,sha256=qA1XbOqYkMRnp29p8VxXjfcH0kHE8NSO5s2heea83-8,21773
 sklearnex/ensemble/__init__.py,sha256=2xB2KfG7l7uJv0p3kfVhrs4KLM174SCsigZhYqwSYAA,1035
-sklearnex/ensemble/_forest.py,sha256=kOtLig8WVipa2VbgMgTQAkHkbSd5UHcz05uMJCtJ8AE,70553
+sklearnex/ensemble/_forest.py,sha256=6vyyQvg60Df-0v9ITIgwrSj2J1RQvGPa2-8Xs5_qXrQ,70003
 sklearnex/ensemble/tests/test_forest.py,sha256=r1GSPQuVVuWYoNB3ZLiGKfMSwq3lp5i6k-Rgr7InonQ,4456
 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/linear_model/__init__.py,sha256=mzZ0EaBhDH66ETNt2vylznSoZbCYexgL2qE_jKppYYc,1144
 sklearnex/linear_model/coordinate_descent.py,sha256=uZOHIKfFlHoMlNXZCh2MAnZE30fRZlmtcF7UsZQ3Vq4,822
-sklearnex/linear_model/linear.py,sha256=ed7pNKKnRkwa-wC0haUCOGHQoPkA4AuFhKNMzmRL6Fw,13832
+sklearnex/linear_model/incremental_linear.py,sha256=26XywxAyyTM6kUCJsuayRtCijKKmSfckiNjrkn5VzHc,13021
+sklearnex/linear_model/linear.py,sha256=2VPcB7SNWlZKK1iHeAQagM2OxBPGFKK4FlY8TjBSZ3U,11162
 sklearnex/linear_model/logistic_path.py,sha256=Nq3JPXSzE4bjamnP3gGQtsMKks2v7s6bSiuYRnFqrRw,849
-sklearnex/linear_model/logistic_regression.py,sha256=KgQ97a7-4ywoRakCUlqPxp-0s4EMZr2XTah-NupKUUY,12990
+sklearnex/linear_model/logistic_regression.py,sha256=6TRaDJ8nIHxjofIfUvmeJOP9_EwmUgVKKK1sxe9vS5A,14171
 sklearnex/linear_model/ridge.py,sha256=0oxlM5McYYvl0KxK9OIGJKM6lOADuFSPTdfx-efJNTI,810
-sklearnex/linear_model/tests/test_linear.py,sha256=li9LLWgap6YCwvNiHQ9yHduvUIsK1lpXfuVNR3dLwig,3200
+sklearnex/linear_model/tests/test_incremental_linear.py,sha256=AfebVKne6fb36Vyr8uPSKYWa9upfPHYZUURHa4Q-mPs,7303
+sklearnex/linear_model/tests/test_linear.py,sha256=nw5EqvF4sAy1dFBQobX0blvHUxZieDbt3wztQcuHrME,4315
 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
@@ -43,19 +47,19 @@ sklearnex/model_selection/__init__.py,sha256=64045Y-nzLDBepO6IRjt88LhL2DM3KdvpCF
 sklearnex/model_selection/split.py,sha256=qjmy8sRf_QEG8LhT0ivn_tICMCYmt7ffZZV1-rLQnko,824
 sklearnex/model_selection/tests/test_model_selection.py,sha256=3kZIq8kba8SW93DBWXupoRymNStR1_mGGYSErQRnwME,1338
 sklearnex/neighbors/__init__.py,sha256=_GMJ2L-6YechRYI2fOFwHjy0kebUncE81z4FmdXUlb8,1078
-sklearnex/neighbors/_lof.py,sha256=5ze0t8_0EXMwsayKux70zJ0Cl4ndDjzv3Fkr2iUXMUc,8109
-sklearnex/neighbors/common.py,sha256=iia-EUIRUIohDCIGpHbVx2PeDlwUvz4Mj1Tn169jidA,10781
-sklearnex/neighbors/knn_classification.py,sha256=eS_uUEdhlBITAUfsaUsAQRh85g01ltsDAug2Rog4gEQ,11161
-sklearnex/neighbors/knn_regression.py,sha256=76eLRvngIy56RyNa4e6-Hz_OqeXiulm6CEkwO7ICuUA,9977
-sklearnex/neighbors/knn_unsupervised.py,sha256=j9YsD5RzawcKWytr7UR3qNlNni5wkAh3vVZWRBSA7sQ,7542
-sklearnex/neighbors/tests/test_neighbors.py,sha256=gcJeJwDoT7wvyvk423k9TDGTGfbXUSTQETUT-VYQ74U,3409
+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_unsupervised.py,sha256=qkj-hSEq8QetYsWlofnik3PpFvo1iBUbIOhywo8wFWk,5362
+sklearnex/neighbors/tests/test_neighbors.py,sha256=yAlMnLt9GrdT6Ceph5B7iFuMJXpDURiHWTE99oO8EDw,3417
 sklearnex/preview/__init__.py,sha256=OXC_k2kudA13rVesXb5ZlEC7_mKUS7uUra_BR-Tin30,780
 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=jSuU8E6r4fdbbBnxBpWp4ybBa62kCnbBx7zDXyUr0Cs,13007
+sklearnex/preview/cluster/k_means.py,sha256=RdF9mwgvZ1xgyeLhCQA2WaPvPsS1ndktXXcaNj6S460,10464
 sklearnex/preview/covariance/__init__.py,sha256=DPYTk8lwcVphtwU8J3CyUYH8Uz6Zr0Uz4S6nn-RY5iM,825
-sklearnex/preview/covariance/covariance.py,sha256=LrNrGP62FEhTSBVaO-EOYAaq-Rszc1VS2B2IqMgh4oo,4938
-sklearnex/preview/covariance/tests/test_covariance.py,sha256=GYI4bMIhGnjmOXpt8J7R0JQmpm9eOvDjIphugRa4kD8,2140
+sklearnex/preview/covariance/covariance.py,sha256=NjRTWLie7UbpfN16wGSxJSRHRhUnvTn03cPPc8PpPd8,5131
+sklearnex/preview/covariance/tests/test_covariance.py,sha256=FOgvhJxFQo9sNG4Uo-ebuw3ZXJ9tSxdk8qKm0ZJzTdc,2487
 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
@@ -74,25 +78,27 @@ 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=Dt1Iyz1g04zOW6hn9cHa9ruzM_MHAIq0ZEEIxh5s7nI,7167
-sklearnex/svm/nusvc.py,sha256=-KMGIanTxVwT9ovvdvGuGBGoMMFQQpAyn1uCoz4CmXA,9150
-sklearnex/svm/nusvr.py,sha256=bkLU8HRlpnnZGxdL6t44Uo-idw1KQOeq2shKBIXf24g,5237
-sklearnex/svm/svc.py,sha256=2wa6gNGMLE1b-sPapP_i89P9fiALDamo8nRIo8H6VS4,10426
-sklearnex/svm/svr.py,sha256=fmYi0dghOmmyFFVI59COX9-tyouQnSDfHIbs8GY8AHs,5241
+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/tests/test_svm.py,sha256=Ru-aGNGCsRJts7SEEYbnKcVqUx-DqPyUtw-hEoMVpW8,4190
-sklearnex/tests/_utils.py,sha256=XYWbUlcdzFDc2DKgHCBdScFrzarsWQrs9GKsTlJsqPs,5059
+sklearnex/tests/_utils.py,sha256=bQR9PJ3GZzp3eZD35pLE8y-ts2a1VGq9ZbwFzh0e5q8,5598
 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=nNVYn6S05q_J-eepNeGDfUKvzZCcHpCnm5jaZhCoPB8,7540
 sklearnex/tests/test_monkeypatch.py,sha256=_9s_4jFvbttOf8uCuBrS4rn_AzQdFlKyRlthnGacUcU,9669
-sklearnex/tests/test_n_jobs_support.py,sha256=ynfCSdCMnR3yEq1YEf_cilVD7zSe0sS-ZQ9jC0hHo8M,3903
+sklearnex/tests/test_n_jobs_support.py,sha256=NziTP4GwZEDoBe1CDvhHZnp8JpwjYQmCNvXEDfS7Wo4,4313
 sklearnex/tests/test_parallel.py,sha256=0zzlh2VJZWcHw5W4QSUthtAygOb6K0Bg1ekNsrdGJQE,1770
-sklearnex/tests/test_patching.py,sha256=eZQRz4GsUkm9yMZj_LgT62hMbUNVdTJg88_UccezoJA,13187
-sklearnex/tests/test_run_to_run_stability_tests.py,sha256=4HDOeJruA1EDILbyQJtsHFmEXC0D1upSHuOT-KyTlEc,14008
-sklearnex/utils/__init__.py,sha256=I8mbJQ3Zsm_F3sCLAhJQb7tUrG30kVsQ-wZoqA8vDdA,842
+sklearnex/tests/test_patching.py,sha256=mDTw-tzO8yibUeRp_ZE560fXnwfnsSq-JAHmXihx1hI,14920
+sklearnex/tests/test_run_to_run_stability_tests.py,sha256=DFKMdIGyzGOBbi7A-o3zPZHLPAIDGIfLRxWzKIYa4hw,14019
+sklearnex/utils/__init__.py,sha256=FFlM8zc2qnSdPYpLjSt4Ma_9FjC8qk8wXJa-5B8hCs0,898
+sklearnex/utils/_namespace.py,sha256=lWzvzcObxq350A-Ms8JyTbaQStR8rgH4DgQYhWruHL4,3166
 sklearnex/utils/parallel.py,sha256=VBcS-KUdyq7XpJUN6ygmNjyWtYLroghbvCxQ8nVU3YI,2085
 sklearnex/utils/validation.py,sha256=fjfhQiKnBQnD7LCBlacMyvsrhGnlMLRXk5Q69uoZIP4,827
-scikit_learn_intelex-2024.3.0.dist-info/LICENSE.txt,sha256=7micbUpzQXphq9e_2oL7PpZcvoXzPuQHIDEXyKXC81s,10797
-scikit_learn_intelex-2024.3.0.dist-info/METADATA,sha256=KeIxHsT7qgO6Z0kvFuwF1SIZ5_fPWgQmgM40Iq7V1BE,12449
-scikit_learn_intelex-2024.3.0.dist-info/WHEEL,sha256=8HLRuLcXtw8zhRBAbl72hZgSmSYmdfyBfCzEjumF0eI,108
-scikit_learn_intelex-2024.3.0.dist-info/top_level.txt,sha256=kzKChSWGJEYFmdj5PwE63HNuP_PVOhWfD32ytH9rL9Q,10
-scikit_learn_intelex-2024.3.0.dist-info/RECORD,,
+sklearnex/utils/tests/test_finite.py,sha256=AzJRY71X0VvDUicUI8Ey9Le6_yKp5O-3ZikhDVJNWms,2943
+scikit_learn_intelex-2024.5.0.dist-info/LICENSE.txt,sha256=7micbUpzQXphq9e_2oL7PpZcvoXzPuQHIDEXyKXC81s,10797
+scikit_learn_intelex-2024.5.0.dist-info/METADATA,sha256=dvtXeQgf40S23qHORmZUGIJaFgl0KF5sqtvHueNvq3Y,12449
+scikit_learn_intelex-2024.5.0.dist-info/WHEEL,sha256=8HLRuLcXtw8zhRBAbl72hZgSmSYmdfyBfCzEjumF0eI,108
+scikit_learn_intelex-2024.5.0.dist-info/top_level.txt,sha256=kzKChSWGJEYFmdj5PwE63HNuP_PVOhWfD32ytH9rL9Q,10
+scikit_learn_intelex-2024.5.0.dist-info/RECORD,,

sklearnex/_device_offload.py

@@ -16,6 +16,7 @@
 
 import logging
 import sys
+from collections.abc import Iterable
 from functools import wraps
 
 import numpy as np
@@ -126,8 +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:
@@ -200,9 +208,35 @@ def _copy_to_usm(queue, array):
         raise RuntimeError(
             "dpctl need to be installed to work " "with __sycl_usm_array_interface__"
         )
-    mem = MemoryUSMDevice(array.nbytes, queue=queue)
-    mem.copy_from_host(array.tobytes())
-    return usm_ndarray(array.shape, array.dtype, buffer=mem)
+
+    if hasattr(array, "__array__"):
+
+        try:
+            mem = MemoryUSMDevice(array.nbytes, queue=queue)
+            mem.copy_from_host(array.tobytes())
+            return usm_ndarray(array.shape, array.dtype, buffer=mem)
+        except ValueError as e:
+            # ValueError will raise if device does not support the dtype
+            # retry with float32 (needed for fp16 and fp64 support issues)
+            # try again as float32, if it is a float32 just raise the error.
+            if array.dtype == np.float32:
+                raise e
+            return _copy_to_usm(queue, array.astype(np.float32))
+    else:
+        if isinstance(array, Iterable):
+            array = [_copy_to_usm(queue, i) for i in array]
+        return array
+
+
+if dpnp_available:
+
+    def _convert_to_dpnp(array):
+        if isinstance(array, usm_ndarray):
+            return dpnp.array(array, copy=False)
+        elif isinstance(array, Iterable):
+            for i in range(len(array)):
+                array[i] = _convert_to_dpnp(array[i])
+        return array
 
 
 def wrap_output_data(func):
@@ -217,7 +251,7 @@ def wrap_output_data(func):
         if usm_iface is not None:
             result = _copy_to_usm(usm_iface["syclobj"], result)
             if dpnp_available and isinstance(data[0], dpnp.ndarray):
-                result = dpnp.array(result, copy=False)
+                result = _convert_to_dpnp(result)
         return result
 
     return wrapper

sklearnex/basic_statistics/__init__.py

@@ -15,5 +15,6 @@
 # ==============================================================================
 
 from .basic_statistics import BasicStatistics
+from .incremental_basic_statistics import IncrementalBasicStatistics
 
-__all__ = ["BasicStatistics"]
+__all__ = ["BasicStatistics", "IncrementalBasicStatistics"]

sklearnex/basic_statistics/incremental_basic_statistics.py

@@ -0,0 +1,288 @@
+# ==============================================================================
+# Copyright 2024 Intel Corporation
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+# ==============================================================================
+
+import numpy as np
+from sklearn.base import BaseEstimator
+from sklearn.utils import check_array, gen_batches
+from sklearn.utils.validation import _check_sample_weight
+
+from daal4py.sklearn._n_jobs_support import control_n_jobs
+from daal4py.sklearn._utils import sklearn_check_version
+from onedal.basic_statistics import (
+    IncrementalBasicStatistics as onedal_IncrementalBasicStatistics,
+)
+
+from .._device_offload import dispatch
+from .._utils import PatchingConditionsChain
+
+if sklearn_check_version("1.2"):
+    from sklearn.utils._param_validation import Interval, StrOptions
+
+import numbers
+
+
+@control_n_jobs(decorated_methods=["partial_fit", "_onedal_finalize_fit"])
+class IncrementalBasicStatistics(BaseEstimator):
+    """
+    Incremental estimator for basic statistics.
+    Allows to compute basic statistics if data are splitted into batches.
+    Parameters
+    ----------
+    result_options: string or list, default='all'
+        List of statistics to compute
+
+    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.
+
+    Attributes (are existing only if corresponding result option exists)
+    ----------
+        min : ndarray of shape (n_features,)
+            Minimum of each feature over all samples.
+
+        max : ndarray of shape (n_features,)
+            Maximum of each feature over all samples.
+
+        sum : ndarray of shape (n_features,)
+            Sum of each feature over all samples.
+
+        mean : ndarray of shape (n_features,)
+            Mean of each feature over all samples.
+
+        variance : ndarray of shape (n_features,)
+            Variance of each feature over all samples.
+
+        variation : ndarray of shape (n_features,)
+            Variation of each feature over all samples.
+
+        sum_squares : ndarray of shape (n_features,)
+            Sum of squares for each feature over all samples.
+
+        standard_deviation : ndarray of shape (n_features,)
+            Standard deviation of each feature over all samples.
+
+        sum_squares_centered : ndarray of shape (n_features,)
+            Centered sum of squares for each feature over all samples.
+
+        second_order_raw_moment : ndarray of shape (n_features,)
+            Second order moment of each feature over all samples.
+    """
+
+    _onedal_incremental_basic_statistics = staticmethod(onedal_IncrementalBasicStatistics)
+
+    if sklearn_check_version("1.2"):
+        _parameter_constraints: dict = {
+            "result_options": [
+                StrOptions(
+                    {
+                        "all",
+                        "min",
+                        "max",
+                        "sum",
+                        "mean",
+                        "variance",
+                        "variation",
+                        "sum_squares",
+                        "standard_deviation",
+                        "sum_squares_centered",
+                        "second_order_raw_moment",
+                    }
+                ),
+                list,
+            ],
+            "batch_size": [Interval(numbers.Integral, 1, None, closed="left"), None],
+        }
+
+    def __init__(self, result_options="all", batch_size=None):
+        if result_options == "all":
+            self.result_options = (
+                self._onedal_incremental_basic_statistics.get_all_result_options()
+            )
+        else:
+            self.result_options = result_options
+        self._need_to_finalize = False
+        self.batch_size = batch_size
+
+    def _onedal_supported(self, method_name, *data):
+        patching_status = PatchingConditionsChain(
+            f"sklearn.covariance.{self.__class__.__name__}.{method_name}"
+        )
+        return patching_status
+
+    _onedal_cpu_supported = _onedal_supported
+    _onedal_gpu_supported = _onedal_supported
+
+    def _get_onedal_result_options(self, options):
+        if isinstance(options, list):
+            onedal_options = "|".join(self.result_options)
+        else:
+            onedal_options = options
+        assert isinstance(onedal_options, str)
+        return options
+
+    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, 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,
+                dtype=[np.float64, np.float32],
+                reset=first_pass,
+            )
+        else:
+            X = check_array(
+                X,
+                dtype=[np.float64, np.float32],
+            )
+
+        if sample_weight is not None:
+            sample_weight = _check_sample_weight(sample_weight, X)
+
+        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]
+
+        onedal_params = {
+            "result_options": self._get_onedal_result_options(self.result_options)
+        }
+        if not hasattr(self, "_onedal_estimator"):
+            self._onedal_estimator = self._onedal_incremental_basic_statistics(
+                **onedal_params
+            )
+        self._onedal_estimator.partial_fit(X, sample_weight, queue)
+        self._need_to_finalize = True
+
+    def _onedal_fit(self, X, sample_weight=None, queue=None):
+        if sklearn_check_version("1.0"):
+            X = self._validate_data(X, dtype=[np.float64, np.float32])
+        else:
+            X = check_array(X, dtype=[np.float64, np.float32])
+
+        if sample_weight is not None:
+            sample_weight = _check_sample_weight(sample_weight, X)
+
+        n_samples, n_features = X.shape
+        if self.batch_size is None:
+            self.batch_size_ = 5 * n_features
+        else:
+            self.batch_size_ = self.batch_size
+
+        self.n_samples_seen_ = 0
+        if hasattr(self, "_onedal_estimator"):
+            self._onedal_estimator._reset()
+
+        for batch in gen_batches(X.shape[0], self.batch_size_):
+            X_batch = X[batch]
+            weights_batch = sample_weight[batch] if sample_weight is not None else None
+            self._onedal_partial_fit(X_batch, weights_batch, queue=queue)
+
+        if sklearn_check_version("1.2"):
+            self._validate_params()
+
+        self.n_features_in_ = X.shape[1]
+
+        self._onedal_finalize_fit()
+
+        return self
+
+    def __getattr__(self, attr):
+        result_options = self.__dict__["result_options"]
+        is_statistic_attr = (
+            isinstance(result_options, str) and (attr == result_options)
+        ) or (isinstance(result_options, list) and (attr in result_options))
+        if is_statistic_attr:
+            if self._need_to_finalize:
+                self._onedal_finalize_fit()
+            return getattr(self._onedal_estimator, attr)
+        if attr in self.__dict__:
+            return self.__dict__[attr]
+
+        raise AttributeError(
+            f"'{self.__class__.__name__}' object has no attribute '{attr}'"
+        )
+
+    def partial_fit(self, X, sample_weight=None):
+        """Incremental fit with X. All of X is processed as a single batch.
+
+        Parameters
+        ----------
+        X : array-like of shape (n_samples, n_features)
+            Data for compute, 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.
+
+        sample_weight : array-like of shape (n_samples,), default=None
+            Weights for compute weighted statistics, where `n_samples` is the number of samples.
+
+        Returns
+        -------
+        self : object
+            Returns the instance itself.
+        """
+        dispatch(
+            self,
+            "partial_fit",
+            {
+                "onedal": self.__class__._onedal_partial_fit,
+                "sklearn": None,
+            },
+            X,
+            sample_weight,
+        )
+        return self
+
+    def fit(self, X, y=None, sample_weight=None):
+        """Compute statistics with X, using minibatches of size batch_size.
+
+        Parameters
+        ----------
+        X : array-like of shape (n_samples, n_features)
+            Data for compute, 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.
+
+        sample_weight : array-like of shape (n_samples,), default=None
+            Weights for compute weighted statistics, where `n_samples` is the number of samples.
+
+        Returns
+        -------
+        self : object
+            Returns the instance itself.
+        """
+        dispatch(
+            self,
+            "fit",
+            {
+                "onedal": self.__class__._onedal_fit,
+                "sklearn": None,
+            },
+            X,
+            sample_weight,
+        )
+        return self