Tags: uxlfoundation/scikit-learn-intelex
Tags
Add array_api_dispatch to spmd validation scope (#2940) (#3098) * Add array_api_dispatch to spmd validation scope * formatting * Combine onedal _device_offload.py change * re-add transform output logic * black * minor restoration * restore host transfer if no queue * add new array api algos and import config_context * debugging * kmeans hotfix * remove debug (cherry picked from commit 29ec7ad) Co-authored-by: ethanglaser <42726565+ethanglaser@users.noreply.github.com>
Add NearestNeighbors SPMD API (#2557) * Add NearestNeighbors SPMD API * black format * extend gold data to have multiple rows per rank * formatting * raw inputs support for kneighbors * Reduce rows of synthetic large test * update search size and only use _spmd_assert_allclose * support empty kneighbors() * Update sklearnex/spmd/neighbors/tests/test_neighbors_spmd.py * address comments
[CI] upgrade dpctl/dpnp versions (#2414) * Update build-and-test-lnx.yml * Update ci.yml * Update build-and-test-lnx.yml * Update renovate.json * Update renovate.json * Update test_common.py * Update test_common.py * Update test_common.py * Update build-and-test-lnx.yml * Update build-and-test-lnx.yml * Update ci.yml * Update renovate.json * Update build-and-test-lnx.yml * Update test_common.py * Update test_common.py * Update common.py * Update coordinate_descent.py * Update logistic_regression.py * Update logistic_regression.py * Update _forest.py * Update _forest.py * Update ci.yml * Update ci.yml * Update _data_conversion.py * Update _data_conversion.py * Update _device_offload.py * try again * fixes * fix KMeans transform issue * formatting * further fixes * Update install_dpcpp.sh * stopping point * Update ci.yml * face facts * clarification * add renovatebot checks * switch source for dpctl/dpnp tracking * fix potential issue with nightly.yml * fix bug * forgotten save * try and fix CI * fix method vs staticmethod issue * try again * try again * move validate_data * Update test_data.py * reformat * try again * try to lazy import changes for dpnp * try with the spelling correction * try again * Update ci.yml * Update _data_conversion.py * Update coordinate_descent.py * Update _data_conversion.py * Update coordinate_descent.py * Update _data_conversion.py * Update ci.yml * Update ci.yml * fix issues related to dpctl/dpnp * further fixes * final hopefully * sometime I can be really dumb * sometime I can be really dumb * Update _device_offload.py * fixes and requests * try to fix spmd use_raw_input issues * fixes and requests
Fixes for sklearn 1.7 pre-release support (#2451) (#2527) * Fixes for sklearn 1.7 pre-release support * Fix ensemble probabilities interval * Change scaling method to `clip` --------- (cherry picked from commit 95c73bd) Co-authored-by: Alexander Andreev <alexander.andreev@intel.com> Co-authored-by: icfaust <icfaust@gmail.com>
ENH: Array API dispatching (#2096) (#2545) * ENH: array api dispatching added array-api-compat to test env * Deselect some scikit-learn Array API tests * deselect more tests * deselect more tests * disabled tests for * fix the deselection comment * disabled test for Ridge regression * Disabled tests and added comment * ENH: Array API dispatching * Revert adding dpctl into Array PI conformance testing added versioning for the get_nnamespace * minor refactoring onedal _array_api * add tests * addressed memory usage tests * Address some array api test fails * linting * addressed test_get_namespace * adding test case for validate_data check with Array API inputs * minor refactoring * addressed test_patch_map_match fail * Added docstrings for get_namespace * docstrings for Array API tests * updated minimal scikit-learn version for Array API dispatching * updated minimal scikit-learn version for Array API dispatching in _device_offload.py _array_api.py * fix test test_get_namespace_with_config_context * refactor onedal/datatypes/_data_conversion.py * correction for array api * Update conftest.py * introduce tags * fix imports * see if this works * really lazy logic introduction * introduce IntelEstimator * missing change in knn * recofigure logic * strip out dpnp/dpctl special code, will come back to it later * switchover * Update __init__.py * Update test_array_api.py * merge main into PR * try to fix changes: * add test: * attempt to get things running * remove tsne failure * try again * add sklearn_check_version * oops * update * another try * try again * Update test_common.py * Update test_common.py * Update test_common.py * Update test_common.py * Update test_common.py * fixes * attempt to re-introduce changes * Update _device_offload.py * switch get_tags * merge main into branch * remove leftover code * Update base.py * formatting * correct bad sklearn recommendation * remove irrelevant new tests * fix mistake * lint fix * add fallback test * make test fixes * formatting * rejigger the logic again * Update test_config.py * interim fix * Update test_config.py * attempt to get test_config test to skip * fix test * interim fixes * fixes for IncrementalEmpiricalCovariance * fixes for wrap_output_data * solve issues related to older numpy * Update test_run_to_run_stability.py * Update __init__.py * Update __init__.py * Update test_run_to_run_stability.py * Update base.py * Update test_run_to_run_stability.py * Update test_config.py * Update test_patching.py * Update test_patching.py * Update test_patching.py * Update k_means.py * Update k_means.py * Update k_means.py * Update _array_api.py * Update _array_api.py * Update _device_offload.py * Update _device_offload.py * Update _device_offload.py * Update _device_offload.py * Update test_patching.py * Update test_patching.py * Update test_patching.py * local verify doclinks * fix doclink * fixes * remove print statement * fix changes from scikit-learn/scikit-learn#29774 * Update sklearnex/_device_offload.py * add type hints and fix docs for dispatch * remove change from local testing * be more explicit with a type * python3.9 fix * change language which was bothering me * formatting * This should fix it (I think) * Apply suggestions from code review * Update base.py * Update __init__.py * remove hack on config * Apply suggestions from code review * make suggested changes * fix import in test_patching * Update test_patching.py * Update build-and-test-lnx.yml * Update build-and-test-lnx.yml * Update build-and-test-lnx.yml * Apply suggestions from code review * Apply suggestions from code review * formatting and ABC change * fix gpu test * remove workaround --------- (cherry picked from commit 43596ee) Co-authored-by: Samir Nasibli <samir.nasibli@intel.com> Co-authored-by: Faust, Ian <ian.faust@intel.com> Co-authored-by: Ian Faust <icfaust@gmail.com> Co-authored-by: david-cortes-intel <david.cortes@intel.com> Co-authored-by: Andreas Huber <9201869+ahuber21@users.noreply.github.com>
PreviousNext