Add integer dtypes (int8, int16, uint8-uint64) to dpnp interface (Int…

…elPython#2230) Add missing aliases to integer types: [u]byte, ushort, [u]longlong Not adding [u]long aliases as it is removed from NumPy (https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations) Removed int alias as it is depricate in NumPy Extend tests types matrix by these types Add environment variable DPNP_TEST_ALL_INT_TYPES to enable/disable these types in tests.
antonwolfy · Feb 14, 2025 · 426da51 · 426da51
1 parent 0403a7c
commit 426da51
Show file tree

Hide file tree

Showing 52 changed files with 1,154 additions and 627 deletions.
diff --git a/.github/workflows/array-api-skips.txt b/.github/workflows/array-api-skips.txt
@@ -1,8 +1,5 @@
 # array API tests to be skipped
 
-# no 'uint8' dtype
-array_api_tests/test_array_object.py::test_getitem_masking
-
 # missing unique-like functions
 array_api_tests/test_has_names.py::test_has_names[set-unique_all]
 array_api_tests/test_has_names.py::test_has_names[set-unique_counts]

diff --git a/.github/workflows/conda-package.yml b/.github/workflows/conda-package.yml
@@ -16,6 +16,7 @@ env:
   CONDA_BUILD_INDEX_ENV_PY_VER: '3.12' # conda does not support python 3.13
   CONDA_BUILD_VERSION: '25.1.1'
   CONDA_INDEX_VERSION: '0.5.0'
+  LATEST_PYTHON: '3.13'
   RERUN_TESTS_ON_FAILURE: 'true'
   RUN_TESTS_MAX_ATTEMPTS: 2
   TEST_ENV_NAME: 'test'
@@ -189,7 +190,7 @@ jobs:
         id: install_dpnp
         continue-on-error: true
         run: |
-          mamba install ${{ env.PACKAGE_NAME }}=${{ env.PACKAGE_VERSION }} pytest python=${{ matrix.python }} ${{ env.TEST_CHANNELS }}
+          mamba install ${{ env.PACKAGE_NAME }}=${{ env.PACKAGE_VERSION }} pytest pytest-xdist python=${{ matrix.python }} ${{ env.TEST_CHANNELS }}
         env:
           TEST_CHANNELS: '-c ${{ env.channel-path }} ${{ env.CHANNELS }}'
 
@@ -211,22 +212,32 @@ jobs:
       - name: Run tests
         if: env.RERUN_TESTS_ON_FAILURE != 'true'
         run: |
-          python -m pytest -ra --pyargs ${{ env.PACKAGE_NAME }}.tests
+          if [[ ${{ matrix.python }} == ${{ env.LATEST_PYTHON }} ]]; then
+            export DPNP_TEST_ALL_INT_TYPES=1
+            python -m pytest -ra --pyargs ${{ env.PACKAGE_NAME }}.tests
+          else
+            python -m pytest -n auto -ra --pyargs ${{ env.PACKAGE_NAME }}.tests
+          fi
 
       - name: Run tests
         if: env.RERUN_TESTS_ON_FAILURE == 'true'
         id: run_tests_linux
         uses: nick-fields/retry@7152eba30c6575329ac0576536151aca5a72780e # v3.0.0
         with:
-          timeout_minutes: 15
+          timeout_minutes: 25
           max_attempts: ${{ env.RUN_TESTS_MAX_ATTEMPTS }}
           retry_on: any
           command: |
             . $CONDA/etc/profile.d/conda.sh
             . $CONDA/etc/profile.d/mamba.sh
             mamba activate ${{ env.TEST_ENV_NAME }}
 
-            python -m pytest -ra --pyargs ${{ env.PACKAGE_NAME }}.tests
+            if [[ ${{ matrix.python }} == ${{ env.LATEST_PYTHON }} ]]; then
+              export DPNP_TEST_ALL_INT_TYPES=1
+              python -m pytest -ra --pyargs ${{ env.PACKAGE_NAME }}.tests
+            else
+              python -m pytest -n auto -ra --pyargs ${{ env.PACKAGE_NAME }}.tests
+            fi
 
   test_windows:
     name: Test
@@ -319,7 +330,7 @@ jobs:
       - name: Install dpnp
         run: |
           @echo on
-          mamba install ${{ env.PACKAGE_NAME }}=${{ env.PACKAGE_VERSION }} pytest python=${{ matrix.python }} ${{ env.TEST_CHANNELS }}
+          mamba install ${{ env.PACKAGE_NAME }}=${{ env.PACKAGE_VERSION }} pytest pytest-xdist python=${{ matrix.python }} ${{ env.TEST_CHANNELS }}
         env:
           TEST_CHANNELS: '-c ${{ env.channel-path }} ${{ env.CHANNELS }}'
           MAMBA_NO_LOW_SPEED_LIMIT: 1
@@ -348,18 +359,28 @@ jobs:
       - name: Run tests
         if: env.RERUN_TESTS_ON_FAILURE != 'true'
         run: |
-          pytest -ra --pyargs ${{ env.PACKAGE_NAME }}.tests
+          if (${{ matrix.python }} -eq ${{ env.LATEST_PYTHON }}) {
+            set DPNP_TEST_ALL_INT_TYPES=1
+            python -m pytest -ra --pyargs ${{ env.PACKAGE_NAME }}.tests
+          } else {
+            python -m pytest -n auto -ra --pyargs ${{ env.PACKAGE_NAME }}.tests
+          }
 
       - name: Run tests
         if: env.RERUN_TESTS_ON_FAILURE == 'true'
         id: run_tests_win
         uses: nick-fields/retry@7152eba30c6575329ac0576536151aca5a72780e # v3.0.0
         with:
-          timeout_minutes: 17
+          timeout_minutes: 35
           max_attempts: ${{ env.RUN_TESTS_MAX_ATTEMPTS }}
           retry_on: any
           command: |
-            python -m pytest -ra --pyargs ${{ env.PACKAGE_NAME }}.tests
+            if ( ${{ matrix.python }} -eq ${{ env.LATEST_PYTHON }} ) {
+              set DPNP_TEST_ALL_INT_TYPES=1
+              python -m pytest -ra --pyargs ${{ env.PACKAGE_NAME }}.tests
+            } else {
+              python -m pytest -n auto -ra --pyargs ${{ env.PACKAGE_NAME }}.tests
+            }
 
   upload:
     name: Upload

diff --git a/doc/reference/dtypes_table.rst b/doc/reference/dtypes_table.rst
@@ -13,22 +13,35 @@ Table below shows a list of all supported data types (dtypes) and constants of t
       - Constants
     * -
         - :obj:`bool <numpy.bool_>`
+        - :obj:`int8 <numpy.int8>`
+        - :obj:`int16 <numpy.int16>`
         - :obj:`int32 <numpy.int32>`
         - :obj:`int64 <numpy.int64>`
+        - :obj:`uint8 <numpy.uint8>`
+        - :obj:`uint16 <numpy.uint16>`
+        - :obj:`uint32 <numpy.uint32>`
+        - :obj:`uint64 <numpy.uint64>`
         - :obj:`float32 <numpy.float32>`
         - :obj:`float64 <numpy.float64>`
         - :obj:`complex64 <numpy.complex64>`
         - :obj:`complex128 <numpy.complex128>`
       -
         - :obj:`bool_ <numpy.bool_>`
+        - :obj:`byte <numpy.byte>`
         - :obj:`cdouble <numpy.cdouble>`
         - :obj:`csingle <numpy.csingle>`
         - :obj:`double <numpy.double>`
         - :obj:`float16 <numpy.float16>`
-        - :obj:`int <numpy.int>`
         - :obj:`int_ <numpy.int_>`
         - :obj:`intc <numpy.intc>`
+        - :obj:`intp <numpy.intp>`
+        - :obj:`longlong <numpy.longlong>`
         - :obj:`single <numpy.single>`
+        - :obj:`ubyte <numpy.ubyte>`
+        - :obj:`uintc <numpy.uintc>`
+        - :obj:`uintp <numpy.uintp>`
+        - :obj:`ushort <numpy.ushort>`
+        - :obj:`ulonglong <numpy.ulonglong>`
       -
         - :obj:`e <numpy.e>`
         - :obj:`euler_gamma <numpy.euler_gamma>`

diff --git a/dpnp/dpnp_algo/dpnp_elementwise_common.py b/dpnp/dpnp_algo/dpnp_elementwise_common.py
@@ -600,12 +600,18 @@ def __init__(
     def __call__(self, x, decimals=0, out=None, dtype=None):
         if decimals != 0:
             x_usm = dpnp.get_usm_ndarray(x)
-            if dpnp.issubdtype(x_usm.dtype, dpnp.integer) and dtype is None:
-                dtype = x_usm.dtype
-
             out_usm = None if out is None else dpnp.get_usm_ndarray(out)
-            x_usm = dpt.round(x_usm * 10**decimals, out=out_usm)
-            res_usm = dpt.divide(x_usm, 10**decimals, out=out_usm)
+
+            if dpnp.issubdtype(x_usm.dtype, dpnp.integer):
+                if decimals < 0:
+                    dtype = x_usm.dtype
+                    x_usm = dpt.round(x_usm * 10**decimals, out=out_usm)
+                    res_usm = dpt.divide(x_usm, 10**decimals, out=out_usm)
+                else:
+                    res_usm = dpt.round(x_usm, out=out_usm)
+            else:
+                x_usm = dpt.round(x_usm * 10**decimals, out=out_usm)
+                res_usm = dpt.divide(x_usm, 10**decimals, out=out_usm)
 
             if dtype is not None:
                 res_usm = dpt.astype(res_usm, dtype, copy=False)

diff --git a/dpnp/dpnp_iface_linearalgebra.py b/dpnp/dpnp_iface_linearalgebra.py
@@ -37,10 +37,12 @@
 
 """
 
+# pylint: disable=no-name-in-module
 import numpy
 
 import dpnp
 
+from .dpnp_utils import map_dtype_to_device
 from .dpnp_utils.dpnp_utils_einsum import dpnp_einsum
 from .dpnp_utils.dpnp_utils_linearalgebra import (
     dpnp_dot,
@@ -66,6 +68,20 @@
 ]
 
 
+# TODO: implement a specific scalar-array kernel
+def _call_multiply(a, b, out=None):
+    """Call multiply function for special cases of scalar-array dots."""
+
+    sc, arr = (a, b) if dpnp.isscalar(a) else (b, a)
+    sc_dtype = map_dtype_to_device(type(sc), arr.sycl_device)
+    res_dtype = dpnp.result_type(sc_dtype, arr)
+    if out is not None and out.dtype == arr.dtype:
+        res = dpnp.multiply(a, b, out=out)
+    else:
+        res = dpnp.multiply(a, b, dtype=res_dtype)
+    return dpnp.get_result_array(res, out, casting="no")
+
+
 def dot(a, b, out=None):
     """
     Dot product of `a` and `b`.
@@ -139,8 +155,7 @@ def dot(a, b, out=None):
             raise ValueError("Only C-contiguous array is acceptable.")
 
     if dpnp.isscalar(a) or dpnp.isscalar(b):
-        # TODO: use specific scalar-vector kernel
-        return dpnp.multiply(a, b, out=out)
+        return _call_multiply(a, b, out=out)
 
     a_ndim = a.ndim
     b_ndim = b.ndim
@@ -635,8 +650,7 @@ def inner(a, b):
     dpnp.check_supported_arrays_type(a, b, scalar_type=True)
 
     if dpnp.isscalar(a) or dpnp.isscalar(b):
-        # TODO: use specific scalar-vector kernel
-        return dpnp.multiply(a, b)
+        return _call_multiply(a, b)
 
     if a.ndim == 0 or b.ndim == 0:
         # TODO: use specific scalar-vector kernel
@@ -714,8 +728,7 @@ def kron(a, b):
     dpnp.check_supported_arrays_type(a, b, scalar_type=True)
 
     if dpnp.isscalar(a) or dpnp.isscalar(b):
-        # TODO: use specific scalar-vector kernel
-        return dpnp.multiply(a, b)
+        return _call_multiply(a, b)
 
     a_ndim = a.ndim
     b_ndim = b.ndim
@@ -1199,8 +1212,7 @@ def tensordot(a, b, axes=2):
             raise ValueError(
                 "One of the inputs is scalar, axes should be zero."
             )
-        # TODO: use specific scalar-vector kernel
-        return dpnp.multiply(a, b)
+        return _call_multiply(a, b)
 
     return dpnp_tensordot(a, b, axes=axes)
 
@@ -1263,13 +1275,13 @@ def vdot(a, b):
         if b.size != 1:
             raise ValueError("The second array should be of size one.")
         a_conj = numpy.conj(a)
-        return dpnp.multiply(a_conj, b)
+        return _call_multiply(a_conj, b)
 
     if dpnp.isscalar(b):
         if a.size != 1:
             raise ValueError("The first array should be of size one.")
         a_conj = dpnp.conj(a)
-        return dpnp.multiply(a_conj, b)
+        return _call_multiply(a_conj, b)
 
     if a.ndim == 1 and b.ndim == 1:
         return dpnp_dot(a, b, out=None, conjugate=True)

diff --git a/dpnp/dpnp_iface_manipulation.py b/dpnp/dpnp_iface_manipulation.py
@@ -1494,7 +1494,20 @@ def copyto(dst, src, casting="same_kind", where=True):
             f"but got {type(dst)}"
         )
     if not dpnp.is_supported_array_type(src):
+        no_dtype_attr = not hasattr(src, "dtype")
         src = dpnp.array(src, sycl_queue=dst.sycl_queue)
+        if no_dtype_attr:
+            # This case (scalar, list, etc) needs special handling to
+            # behave similar to NumPy
+            if dpnp.issubdtype(src, dpnp.integer) and dpnp.issubdtype(
+                dst, dpnp.unsignedinteger
+            ):
+                if dpnp.any(src < 0):
+                    raise OverflowError(
+                        "Cannot copy negative values to an unsigned int array"
+                    )
+
+                src = src.astype(dst.dtype)
 
     if not dpnp.can_cast(src.dtype, dst.dtype, casting=casting):
         raise TypeError(

diff --git a/dpnp/dpnp_iface_types.py b/dpnp/dpnp_iface_types.py
@@ -40,6 +40,7 @@
 __all__ = [
     "bool",
     "bool_",
+    "byte",
     "cdouble",
     "complex128",
     "complex64",
@@ -57,8 +58,9 @@
     "iinfo",
     "inexact",
     "inf",
-    "int",
     "int_",
+    "int8",
+    "int16",
     "int32",
     "int64",
     "integer",
@@ -67,12 +69,24 @@
     "isdtype",
     "issubdtype",
     "is_type_supported",
+    "longlong",
     "nan",
     "newaxis",
     "number",
     "pi",
+    "short",
     "signedinteger",
     "single",
+    "ubyte",
+    "uint8",
+    "uint16",
+    "uint32",
+    "uint64",
+    "uintc",
+    "uintp",
+    "unsignedinteger",
+    "ushort",
+    "ulonglong",
 ]
 
 
@@ -82,6 +96,7 @@
 # =============================================================================
 bool = numpy.bool_
 bool_ = numpy.bool_
+byte = numpy.byte
 cdouble = numpy.cdouble
 complex128 = numpy.complex128
 complex64 = numpy.complex64
@@ -94,16 +109,29 @@
 float64 = numpy.float64
 floating = numpy.floating
 inexact = numpy.inexact
-int = numpy.int_
 int_ = numpy.int_
+int8 = numpy.int8
+int16 = numpy.int16
 int32 = numpy.int32
 int64 = numpy.int64
 integer = numpy.integer
 intc = numpy.intc
 intp = numpy.intp
+longlong = numpy.longlong
 number = numpy.number
+short = numpy.short
 signedinteger = numpy.signedinteger
 single = numpy.single
+ubyte = numpy.ubyte
+uint8 = numpy.uint8
+uint16 = numpy.uint16
+uint32 = numpy.uint32
+uint64 = numpy.uint64
+uintc = numpy.uintc
+uintp = numpy.uintp
+unsignedinteger = numpy.unsignedinteger
+ushort = numpy.ushort
+ulonglong = numpy.ulonglong
 
 
 # =============================================================================