Merge master into impl_lu_solve_2D

Author: vlad-perevezentsev

.github/workflows/check-onemath.yaml

@@ -15,6 +15,7 @@ env:
   build-with-oneapi-env: 'environments/build_with_oneapi.yml'
   dpctl-pkg-env: 'environments/dpctl_pkg.yml'
   oneapi-pkgs-env: 'environments/oneapi_pkgs.yml'
+  test-pkg-env: 'environments/test.yml'
   test-env-name: 'test_onemath'
   rerun-tests-on-failure: 'true'
   rerun-tests-max-attempts: 2
@@ -49,7 +50,10 @@ jobs:
 
       - name: Merge conda env files
         run: |
-          conda-merge ${{ env.dpctl-pkg-env }} ${{ env.oneapi-pkgs-env }} ${{ env.build-with-oneapi-env }} > ${{ env.environment-file }}
+          conda-merge ${{ env.dpctl-pkg-env }}         \
+                      ${{ env.oneapi-pkgs-env }}       \
+                      ${{ env.build-with-oneapi-env }} \
+                      ${{ env.test-pkg-env }} > ${{ env.environment-file }}
           cat ${{ env.environment-file }}
 
       - name: Upload artifact

.github/workflows/cron-run-tests.yaml

@@ -84,12 +84,12 @@ jobs:
         id: install_dpnp
         continue-on-error: true
         run: |
-          mamba install ${{ env.package-name }}=${{ steps.find_latest_tag.outputs.tag }} pytest ${{ env.channels-list }}
+          mamba install ${{ env.package-name }}=${{ steps.find_latest_tag.outputs.tag }} pytest scipy ${{ env.channels-list }}
 
       - name: ReInstall dpnp
         if: steps.install_dpnp.outcome == 'failure'
         run: |
-          mamba install ${{ env.package-name }}=${{ steps.find_latest_tag.outputs.tag }} pytest ${{ env.channels-list }}
+          mamba install ${{ env.package-name }}=${{ steps.find_latest_tag.outputs.tag }} pytest scipy ${{ env.channels-list }}
 
       - name: List installed packages
         run: mamba list

.github/workflows/generate_coverage.yaml

@@ -25,6 +25,7 @@ jobs:
       environment-file: 'environments/environment.yml'
       build-with-oneapi-env: 'environments/build_with_oneapi.yml'
       coverage-env: 'environments/coverage.yml'
+      test-pkg-env: 'environments/test.yml'
       oneapi-pkgs-env: ''
       # Enable env when it's required to use only conda packages without OneAPI installation
       # oneapi-pkgs-env: '${{ github.workspace }}/environments/oneapi_pkgs.yml'
@@ -73,7 +74,10 @@ jobs:
 
       - name: Merge conda env files
         run: |
-          conda-merge ${{ env.build-with-oneapi-env }} ${{ env.coverage-env }} ${{ env.oneapi-pkgs-env }} > ${{ env.environment-file }}
+          conda-merge ${{ env.build-with-oneapi-env }} \
+                      ${{ env.coverage-env }}          \
+                      ${{ env.oneapi-pkgs-env }}       \
+                      ${{ env.test-pkg-env }} > ${{ env.environment-file }}
           cat ${{ env.environment-file }}
 
       - name: Setup miniconda

.pre-commit-config.yaml

@@ -113,7 +113,7 @@ repos:
             "--disable=redefined-builtin",
             "--disable=unused-wildcard-import"
             ]
-        files: '^dpnp/(dpnp_iface.*|fft|linalg|dpnp_array)'
+        files: '^dpnp/(dpnp_iface.*|fft|linalg|special|dpnp_array)'
 -   repo: https://github.com/macisamuele/language-formatters-pre-commit-hooks
     rev: v2.15.0
     hooks:

CHANGELOG.md

@@ -29,7 +29,7 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 * Aligned the license expression with `PEP-639` [#2511](https://github.com/IntelPython/dpnp/pull/2511)
 * Bumped oneMKL version up to `v0.8` [#2514](https://github.com/IntelPython/dpnp/pull/2514)
 * Removed the use of class template argument deduction for alias template to conform to the C++17 standard [#2517](https://github.com/IntelPython/dpnp/pull/2517)
-* Changed th order of individual FFTs over `axes` for `dpnp.fft.irfftn` to be in forward order [#2524](https://github.com/IntelPython/dpnp/pull/2524)
+* Changed the order of individual FFTs over `axes` for `dpnp.fft.irfftn` to be in forward order [#2524](https://github.com/IntelPython/dpnp/pull/2524)
 * Replaced the use of `numpy.testing.suppress_warnings` with appropriate calls from the warnings module [#2529](https://github.com/IntelPython/dpnp/pull/2529)
 * Improved documentations of `dpnp.ndarray` class and added a page with description of supported constants [#2422](https://github.com/IntelPython/dpnp/pull/2422)
 * Updated `dpnp.size` to accept tuple of ints for `axes` argument [#2536](https://github.com/IntelPython/dpnp/pull/2536)
@@ -41,6 +41,8 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 * Refactored backend implementation of `dpnp.linalg.solve` to use oneMKL LAPACK `gesv` directly [#2558](https://github.com/IntelPython/dpnp/pull/2558)
 * Improved performance of `dpnp.isclose` function by implementing a dedicated kernel for scalar `rtol` and `atol` arguments [#2540](https://github.com/IntelPython/dpnp/pull/2540)
 * Extended `dpnp.pad` to support `pad_width` keyword as a dictionary [#2535](https://github.com/IntelPython/dpnp/pull/2535)
+* Redesigned `dpnp.erf` function through pybind11 extension of OneMKL call or dedicated kernel in `ufunc` namespace [#2551](https://github.com/IntelPython/dpnp/pull/2551)
+* Improved performance of batched implementation of `dpnp.linalg.det` and `dpnp.linalg.slogdet` [#2572](https://github.com/IntelPython/dpnp/pull/2572)
 
 ### Deprecated
 

conda-recipe/meta.yaml

@@ -63,6 +63,7 @@ test:
     requires:
       - pytest
       - setuptools
+      - scipy
 
 about:
     home: https://github.com/IntelPython/dpnp

doc/reference/special.rst

@@ -1,17 +1,20 @@
-Special Functions
-=================
+.. currentmodule:: dpnp.special
 
-.. https://docs.scipy.org/doc/scipy/reference/special.html
+Special functions (:mod:`dpnp.special`)
+=======================================
 
-Error Function
---------------
+.. Hint:: `SciPy API Reference: Special functions (scipy.special) <https://docs.scipy.org/doc/scipy/reference/special.html>`_
+
+Error function and Fresnel integrals
+------------------------------------
 
 .. autosummary::
    :toctree: generated/
    :nosignatures:
 
-   dpnp.erf
-   dpnp.erfc
-   dpnp.erfcx
-   dpnp.erfinv
-   dpnp.erfcinv
+   erf
+   erfc
+   erfcx
+   erfi
+   erfinv
+   erfcinv

dpnp/backend/CMakeLists.txt

@@ -26,7 +26,6 @@
 set(DPNP_SRC
     kernels/dpnp_krnl_arraycreation.cpp
     kernels/dpnp_krnl_common.cpp
-    kernels/dpnp_krnl_elemwise.cpp
     kernels/dpnp_krnl_mathematical.cpp
     kernels/dpnp_krnl_random.cpp
     kernels/dpnp_krnl_sorting.cpp

dpnp/backend/extensions/ufunc/CMakeLists.txt

@@ -27,6 +27,7 @@ set(_elementwise_sources
     ${CMAKE_CURRENT_SOURCE_DIR}/elementwise_functions/bitwise_count.cpp
     ${CMAKE_CURRENT_SOURCE_DIR}/elementwise_functions/common.cpp
     ${CMAKE_CURRENT_SOURCE_DIR}/elementwise_functions/degrees.cpp
+    ${CMAKE_CURRENT_SOURCE_DIR}/elementwise_functions/erf.cpp
     ${CMAKE_CURRENT_SOURCE_DIR}/elementwise_functions/fabs.cpp
     ${CMAKE_CURRENT_SOURCE_DIR}/elementwise_functions/fix.cpp
     ${CMAKE_CURRENT_SOURCE_DIR}/elementwise_functions/float_power.cpp

dpnp/backend/extensions/ufunc/elementwise_functions/common.cpp

@@ -27,6 +27,7 @@
 
 #include "bitwise_count.hpp"
 #include "degrees.hpp"
+#include "erf.hpp"
 #include "fabs.hpp"
 #include "fix.hpp"
 #include "float_power.hpp"
@@ -57,6 +58,7 @@ void init_elementwise_functions(py::module_ m)
 {
     init_bitwise_count(m);
     init_degrees(m);
+    init_erf(m);
     init_fabs(m);
     init_fix(m);
     init_float_power(m);