Add dask to array-api-compat

.github/workflows/array-api-tests-dask.yml

@@ -0,0 +1,9 @@
+name: Array API Tests (Dask)
+
+on: [push, pull_request]
+
+jobs:
+  array-api-tests-dask:
+    uses: ./.github/workflows/array-api-tests.yml
+    with:
+      package-name: dask

array_api_compat/common/_aliases.py

@@ -303,6 +303,8 @@ def _asarray(
         import numpy as xp
     elif namespace == 'cupy':
         import cupy as xp
+    elif namespace == 'dask':
+        import dask.array as xp
     else:
         raise ValueError("Unrecognized namespace argument to asarray()")
 
@@ -322,7 +324,9 @@ def _asarray(
     if copy in COPY_FALSE:
         # copy=False is not yet implemented in xp.asarray
         raise NotImplementedError("copy=False is not yet implemented")
-    if isinstance(obj, xp.ndarray):
+    # TODO: This feels wrong (__array__ is not in the standard)
+    # Dask doesn't support DLPack, though, so, this'll do
+    if (hasattr(xp, "ndarray") and isinstance(obj, xp.ndarray)) or hasattr(obj, "__array__"):
         if dtype is not None and obj.dtype != dtype:
             copy = True
         if copy in COPY_TRUE:

array_api_compat/common/_helpers.py

@@ -40,6 +40,16 @@ def _is_torch_array(x):
     # TODO: Should we reject ndarray subclasses?
     return isinstance(x, torch.Tensor)
 
+def _is_dask_array(x):
+    # Avoid importing dask if it isn't already
+    if 'dask.array' not in sys.modules:
+        return False
+
+    import dask.array
+
+    # TODO: Should we reject ndarray subclasses?
+    return isinstance(x, dask.array.Array)
+
 def is_array_api_obj(x):
     """
     Check if x is an array API compatible array object.
@@ -97,6 +107,13 @@ def your_function(x, y):
             else:
                 import torch
                 namespaces.add(torch)
+        elif _is_dask_array(x):
+            _check_api_version(api_version)
+            if _use_compat:
+                from .. import dask as dask_namespace
+                namespaces.add(dask_namespace)
+            else:
+                raise TypeError("_use_compat cannot be False if input array is a dask array!")
         else:
             # TODO: Support Python scalars?
             raise TypeError("The input is not a supported array type")
@@ -219,6 +236,13 @@ def to_device(x: "Array", device: "Device", /, *, stream: "Optional[Union[int, A
         return _cupy_to_device(x, device, stream=stream)
     elif _is_torch_array(x):
         return _torch_to_device(x, device, stream=stream)
+    elif _is_dask_array(x):
+        if stream is not None:
+            raise ValueError("The stream argument to to_device() is not supported")
+        # TODO: What if our array is on the GPU already?
+        if device == 'cpu':
+            return x
+        raise ValueError(f"Unsupported device {device!r}")
     return x.to_device(device, stream=stream)
 
 def size(x):

array_api_compat/dask/__init__.py

@@ -0,0 +1,6 @@
+from dask.array import *
+
+# These imports may overwrite names from the import * above.
+from ._aliases import *
+
+__array_api_version__ = '2022.12'

array_api_compat/dask/_aliases.py

@@ -0,0 +1,88 @@
+from ..common import _aliases
+
+from .._internal import get_xp
+
+import numpy as np
+from numpy import (
+    # Constants
+    e,
+    inf,
+    nan,
+    pi,
+    newaxis,
+    # Dtypes
+    bool_ as bool,
+    float32,
+    float64,
+    int8,
+    int16,
+    int32,
+    int64,
+    uint8,
+    uint16,
+    uint32,
+    uint64,
+    complex64,
+    complex128,
+    iinfo,
+    finfo,
+    can_cast,
+    result_type,
+)
+
+import dask.array as da
+
+isdtype = get_xp(np)(_aliases.isdtype)
+astype = _aliases.astype
+
+# Common aliases
+arange = get_xp(da)(_aliases.arange)
+
+from functools import partial
+asarray = partial(_aliases._asarray, namespace='dask')
+asarray.__doc__ = _aliases._asarray.__doc__
+
+linspace = get_xp(da)(_aliases.linspace)
+eye = get_xp(da)(_aliases.eye)
+UniqueAllResult = get_xp(da)(_aliases.UniqueAllResult)
+UniqueCountsResult = get_xp(da)(_aliases.UniqueCountsResult)
+UniqueInverseResult = get_xp(da)(_aliases.UniqueInverseResult)
+unique_all = get_xp(da)(_aliases.unique_all)
+unique_counts = get_xp(da)(_aliases.unique_counts)
+unique_inverse = get_xp(da)(_aliases.unique_inverse)
+unique_values = get_xp(da)(_aliases.unique_values)
+permute_dims = get_xp(da)(_aliases.permute_dims)
+std = get_xp(da)(_aliases.std)
+var = get_xp(da)(_aliases.var)
+empty = get_xp(da)(_aliases.empty)
+empty_like = get_xp(da)(_aliases.empty_like)
+full = get_xp(da)(_aliases.full)
+full_like = get_xp(da)(_aliases.full_like)
+ones = get_xp(da)(_aliases.ones)
+ones_like = get_xp(da)(_aliases.ones_like)
+zeros = get_xp(da)(_aliases.zeros)
+zeros_like = get_xp(da)(_aliases.zeros_like)
+reshape = get_xp(da)(_aliases.reshape)
+matrix_transpose = get_xp(da)(_aliases.matrix_transpose)
+vecdot = get_xp(da)(_aliases.vecdot)
+
+
+
+from dask.array import (
+    # Element wise aliases
+    arccos as acos,
+    arccosh as acosh,
+    arcsin as asin,
+    arcsinh as asinh,
+    arctan as atan,
+    arctan2 as atan2,
+    arctanh as atanh,
+    left_shift as bitwise_left_shift,
+    right_shift as bitwise_right_shift,
+    invert as bitwise_invert,
+    power as pow,
+    # Other
+    concatenate as concat,
+
+)
+

dask-xfails.txt

@@ -0,0 +1,47 @@
+# finfo(float32).eps returns float32 but should return float
+array_api_tests/test_data_type_functions.py::test_finfo[float32]
+
+# No sorting in dask
+array_api_tests/test_has_names.py::test_has_names[sorting-argsort]
+array_api_tests/test_has_names.py::test_has_names[sorting-sort]
+
+# Array methods and attributes not already on np.ndarray cannot be wrapped
+array_api_tests/test_has_names.py::test_has_names[array_method-__array_namespace__]
+array_api_tests/test_has_names.py::test_has_names[array_method-to_device]
+array_api_tests/test_has_names.py::test_has_names[array_attribute-device]
+array_api_tests/test_has_names.py::test_has_names[array_attribute-mT]
+
+#FAILED array_api_tests/test_operators_and_elementwise_functions.py::test_add[__add__(x1, x2)] - AssertionError: out[0]=0, but should be (x1 + x2[0])=65536 [__add__()]
+#FAILED array_api_tests/test_operators_and_elementwise_functions.py::test_bitwise_and[__and__(x1, x2)] - hypothesis.errors.Flaky: Hypothesis test_bitwise_and(ctx=BinaryParamContext(<__and__(x1, x2)>), data=data(...)) produces unreliable results: Falsified on the ...
+#FAILED array_api_tests/test_operators_and_elementwise_functions.py::test_bitwise_right_shift[__rshift__(x1, x2)] - hypothesis.errors.Flaky: Hypothesis test_bitwise_right_shift(ctx=BinaryParamContext(<__rshift__(x1, x2)>), data=data(...)) produces unreliable results: Falsif...
+#FAILED array_api_tests/test_operators_and_elementwise_functions.py::test_bitwise_xor[bitwise_xor(x1, x2)] - hypothesis.errors.DeadlineExceeded: Test took 994.44ms, which exceeds the deadline of 800.00ms
+#FAILED array_api_tests/test_operators_and_elementwise_functions.py::test_bitwise_xor[__xor__(x1, x2)] - ValueError: Inferred dtype from function 'xor' was 'uint64' but got 'int16', which can't be cast using casting='same_kind'
+#FAILED array_api_tests/test_operators_and_elementwise_functions.py::test_ceil - exceptiongroup.ExceptionGroup: Hypothesis found 2 distinct failures. (2 sub-exceptions)
+#FAILED array_api_tests/test_operators_and_elementwise_functions.py::test_divide[divide(x1, x2)] - hypothesis.errors.Flaky: Hypothesis test_divide(ctx=BinaryParamContext(<divide(x1, x2)>), data=data(...)) produces unreliable results: Falsified on the first ...
+#FAILED array_api_tests/test_operators_and_elementwise_functions.py::test_floor - exceptiongroup.ExceptionGroup: Hypothesis found 2 distinct failures. (2 sub-exceptions)
+#FAILED array_api_tests/test_operators_and_elementwise_functions.py::test_floor_divide[floor_divide(x1, x2)] - hypothesis.errors.DeadlineExceeded: Test took 1015.43ms, which exceeds the deadline of 800.00ms
+#FAILED array_api_tests/test_operators_and_elementwise_functions.py::test_floor_divide[__floordiv__(x1, x2)] - hypothesis.errors.DeadlineExceeded: Test took 900.99ms, which exceeds the deadline of 800.00ms
+#FAILED array_api_tests/test_operators_and_elementwise_functions.py::test_greater[greater(x1, x2)] - hypothesis.errors.DeadlineExceeded: Test took 1106.49ms, which exceeds the deadline of 800.00ms
+#FAILED array_api_tests/test_operators_and_elementwise_functions.py::test_greater[__gt__(x1, x2)] - hypothesis.errors.Flaky: Hypothesis test_greater(ctx=BinaryParamContext(<__gt__(x1, x2)>), data=data(...)) produces unreliable results: Falsified on the first...
+#FAILED array_api_tests/test_operators_and_elementwise_functions.py::test_greater_equal[greater_equal(x1, x2)] - hypothesis.errors.Flaky: Hypothesis test_greater_equal(ctx=BinaryParamContext(<greater_equal(x1, x2)>), data=data(...)) produces unreliable results: Falsified...
+#FAILED array_api_tests/test_operators_and_elementwise_functions.py::test_greater_equal[__ge__(x1, x2)] - hypothesis.errors.Flaky: Hypothesis test_greater_equal(ctx=BinaryParamContext(<__ge__(x1, x2)>), data=data(...)) produces unreliable results: Falsified on the...
+#FAILED array_api_tests/test_operators_and_elementwise_functions.py::test_less[less(x1, x2)] - hypothesis.errors.DeadlineExceeded: Test took 961.84ms, which exceeds the deadline of 800.00ms
+#FAILED array_api_tests/test_operators_and_elementwise_functions.py::test_less[__lt__(x1, x2)] - hypothesis.errors.DeadlineExceeded: Test took 980.68ms, which exceeds the deadline of 800.00ms
+#FAILED array_api_tests/test_operators_and_elementwise_functions.py::test_less_equal[less_equal(x1, x2)] - hypothesis.errors.DeadlineExceeded: Test took 1043.47ms, which exceeds the deadline of 800.00ms
+#FAILED array_api_tests/test_operators_and_elementwise_functions.py::test_less_equal[__le__(x1, x2)] - hypothesis.errors.DeadlineExceeded: Test took 1011.76ms, which exceeds the deadline of 800.00ms
+#FAILED array_api_tests/test_operators_and_elementwise_functions.py::test_multiply[multiply(x1, x2)] - AssertionError: out[0]=0, but should be (x1 * x2[0])=256 [multiply()]
+#FAILED array_api_tests/test_operators_and_elementwise_functions.py::test_multiply[__mul__(x1, x2)] - AssertionError: out[0]=2, but should be (x1 * x2[0])=258 [__mul__()]
+#FAILED array_api_tests/test_operators_and_elementwise_functions.py::test_subtract[subtract(x1, x2)] - exceptiongroup.ExceptionGroup: Hypothesis found 2 distinct failures. (2 sub-exceptions)
+#FAILED array_api_tests/test_operators_and_elementwise_functions.py::test_subtract[__sub__(x1, x2)] - hypothesis.errors.DeadlineExceeded: Test took 1034.64ms, which exceeds the deadline of 800.00ms
+#FAILED array_api_tests/test_operators_and_elementwise_functions.py::test_trunc - exceptiongroup.ExceptionGroup: Hypothesis found 2 distinct failures. (2 sub-exceptions)
+
+#FAILED array_api_tests/test_searching_functions.py::test_nonzero_zerodim_error - Failed: DID NOT RAISE <class 'Exception'>
+
+# Fails because shape is NaN since we don't materialize it yet
+#FAILED array_api_tests/test_searching_functions.py::test_nonzero - AssertionError: prod(out[0].shape)=nan, but should be prod(out[0].shape)=nan
+#FAILED array_api_tests/test_set_functions.py::test_unique_all - AssertionError: out.indices.shape=(nan,), but should be out.values.shape=(nan,)
+#FAILED array_api_tests/test_set_functions.py::test_unique_counts - AssertionError: out.counts.shape=(nan,), but should be out.values.shape=(nan,)
+
+# Needs investigation
+#FAILED array_api_tests/test_set_functions.py::test_unique_inverse - TypeError: 'float' object cannot be interpreted as an integer
+#FAILED array_api_tests/test_set_functions.py::test_unique_values - TypeError: 'float' object cannot be interpreted as an integer

tests/test_array_namespace.py

@@ -5,8 +5,7 @@
 
 import pytest
 
-
-@pytest.mark.parametrize("library", ["cupy", "numpy", "torch"])
+@pytest.mark.parametrize("library", ["cupy", "numpy", "torch", "dask.array"])
 @pytest.mark.parametrize("api_version", [None, '2021.12'])
 def test_array_namespace(library, api_version):
     lib = import_(library)
@@ -17,6 +16,8 @@ def test_array_namespace(library, api_version):
     if 'array_api' in library:
         assert namespace == lib
     else:
+        if library == "dask.array":
+            library = "dask"
         assert namespace == getattr(array_api_compat, library)
 
 

tests/test_common.py

@@ -5,7 +5,7 @@
 import numpy as np
 from numpy.testing import assert_allclose
 
-@pytest.mark.parametrize("library", ["cupy", "numpy", "torch"])
+@pytest.mark.parametrize("library", ["cupy", "numpy", "torch", "dask"])
 def test_to_device_host(library):
     # different libraries have different semantics
     # for DtoH transfers; ensure that we support a portable

tests/test_isdtype.py

@@ -64,7 +64,7 @@ def isdtype_(dtype_, kind):
     assert type(res) is bool
     return res
 
-@pytest.mark.parametrize("library", ["cupy", "numpy", "torch"])
+@pytest.mark.parametrize("library", ["cupy", "numpy", "torch", "dask"])
 def test_isdtype_spec_dtypes(library):
     xp = import_('array_api_compat.' + library)
 
@@ -98,7 +98,7 @@ def test_isdtype_spec_dtypes(library):
     'bfloat16',
 ]
 
-@pytest.mark.parametrize("library", ["cupy", "numpy", "torch"])
+@pytest.mark.parametrize("library", ["cupy", "numpy", "torch", "dask"])
 @pytest.mark.parametrize("dtype_", additional_dtypes)
 def test_isdtype_additional_dtypes(library, dtype_):
     xp = import_('array_api_compat.' + library)

tests/test_vendoring.py

@@ -17,3 +17,7 @@ def test_vendoring_cupy():
 def test_vendoring_torch():
     from vendor_test import uses_torch
     uses_torch._test_torch()
+
+def test_vendoring_torch():
+    from vendor_test import uses_torch
+    uses_torch._test_torch()

vendor_test/uses_dask.py

@@ -0,0 +1,19 @@
+# Basic test that vendoring works
+
+from .vendored._compat import dask as dask_compat
+
+import dask.array as da
+import numpy as np
+
+def _test_numpy():
+    a = dask_compat.asarray([1., 2., 3.])
+    b = dask_compat.arange(3, dtype=dask_compat.float32)
+
+    # np.pow does not exist. Update this to use something else if it is added
+    res = dask_compat.pow(a, b)
+    assert res.dtype == dask_compat.float64 == np.float64
+    assert isinstance(a, da.array)
+    assert isinstance(b, da.array)
+    assert isinstance(res, da.array)
+
+    np.testing.assert_allclose(res, [1., 2., 9.])