diff --git a/test/test_arraycontext.py b/test/test_arraycontext.py
index 752f69bff6acf22b01682ff7e8a2861ee5077952..a8bdcb79377e9b262db0f5ef2597ad94a6a03aa5 100644
--- a/test/test_arraycontext.py
+++ b/test/test_arraycontext.py
@@ -34,7 +34,8 @@ from arraycontext import (
FirstAxisIsElementsTag)
from arraycontext import ( # noqa: F401
pytest_generate_tests_for_pyopencl_array_context
- as pytest_generate_tests)
+ as pytest_generate_tests,
+ _acf)
import logging
logger = logging.getLogger(__name__)
@@ -125,94 +126,94 @@ def _thaw_dofarray(ary, actx):
# }}}
-def test_array_context_np_workalike(actx_factory):
- actx = actx_factory()
+# {{{ assert_close_to_numpy*
- ndofs = 5000
+def assert_close_to_numpy(actx, op, args):
+ assert np.allclose(
+ actx.to_numpy(
+ op(actx.np, *[
+ actx.from_numpy(arg) if isinstance(arg, np.ndarray) else arg
+ for arg in args])),
+ op(np, *args))
- for sym_name, n_args in [
- ("sin", 1),
- ("exp", 1),
- ("arctan2", 2),
- ("minimum", 2),
- ("maximum", 2),
- ("where", 3),
- ("conj", 1),
- # ("empty_like", 1), # NOTE: fails np.allclose, obviously
- ("zeros_like", 1),
- ("ones_like", 1),
- ]:
- args = [np.random.randn(ndofs) for i in range(n_args)]
- ref_result = getattr(np, sym_name)(*args)
- # {{{ test cl.Arrays
+def assert_close_to_numpy_in_containers(actx, op, args):
+ assert_close_to_numpy(actx, op, args)
- actx_args = [actx.from_numpy(arg) for arg in args]
- actx_result = actx.to_numpy(getattr(actx.np, sym_name)(*actx_args))
+ ref_result = op(np, *args)
- assert np.allclose(actx_result, ref_result)
-
- # }}}
-
- # {{{ test DOFArrays
+ # {{{ test DOFArrays
- actx_args = [DOFArray(actx, (arg,)) for arg in actx_args]
- actx_result = actx.to_numpy(
- getattr(actx.np, sym_name)(*actx_args)[0])
+ dofarray_args = [
+ DOFArray(actx, (actx.from_numpy(arg),))
+ if isinstance(arg, np.ndarray) else arg
+ for arg in args]
+ actx_result = actx.to_numpy(op(actx.np, *dofarray_args)[0])
- assert np.allclose(actx_result, ref_result)
+ assert np.allclose(actx_result, ref_result)
- # }}}
+ # }}}
- # {{{ test object arrays
+ # {{{ test object arrays of DOFArrays
- if sym_name in ("empty_like", "zeros_like", "ones_like"):
- continue
+ obj_array_args = [
+ make_obj_array([arg]) if isinstance(arg, DOFArray) else arg
+ for arg in dofarray_args]
+ obj_array_result = actx.to_numpy(op(actx.np, *obj_array_args)[0][0])
- obj_array_args = [make_obj_array([arg]) for arg in actx_args]
- obj_array_result = actx.to_numpy(
- getattr(actx.np, sym_name)(*obj_array_args)[0][0])
+ assert np.allclose(obj_array_result, ref_result)
- assert np.allclose(obj_array_result, ref_result)
+ # }}}
- # }}}
+# }}}
-# {{{ Array manipulations
+# {{{ np.function same as numpy
-def test_actx_stack(actx_factory):
+@pytest.mark.parametrize(("sym_name", "n_args"), [
+ ("sin", 1),
+ ("exp", 1),
+ ("arctan2", 2),
+ ("minimum", 2),
+ ("maximum", 2),
+ ("where", 3),
+ ("conj", 1),
+ ])
+def test_array_context_np_workalike(actx_factory, sym_name, n_args):
actx = actx_factory()
ndofs = 5000
- args = [np.random.randn(ndofs) for i in range(10)]
- ref_result = np.stack(args)
-
- # {{{ test cl.Arrays
+ args = [np.random.randn(ndofs) for i in range(n_args)]
+ assert_close_to_numpy_in_containers(
+ actx, lambda _np, *_args: getattr(_np, sym_name)(*_args), args)
- actx_args = [actx.from_numpy(arg) for arg in args]
- actx_result = actx.to_numpy(actx.np.stack(actx_args))
-
- assert np.allclose(actx_result, ref_result)
- # }}}
+@pytest.mark.parametrize(("sym_name", "n_args"), [
+ # ("empty_like", 1), # NOTE: fails np.allclose, obviously
+ ("zeros_like", 1),
+ ("ones_like", 1),
+ ])
+def test_array_context_np_like(actx_factory, sym_name, n_args):
+ actx = actx_factory()
- # {{{ test DOFArrays
+ ndofs = 5000
+ args = [np.random.randn(ndofs) for i in range(n_args)]
+ assert_close_to_numpy(
+ actx, lambda _np, *_args: getattr(_np, sym_name)(*_args), args)
- actx_args = [DOFArray(actx, (arg,)) for arg in actx_args]
- actx_result = actx.to_numpy(actx.np.stack(actx_args)[0])
+# }}}
- assert np.allclose(actx_result, ref_result)
- # }}}
+# {{{ Array manipulations
- # {{{ test object arrays
+def test_actx_stack(actx_factory):
+ actx = actx_factory()
- obj_array_args = [make_obj_array([arg]) for arg in actx_args]
- obj_array_result = actx.to_numpy(actx.np.stack(obj_array_args)[0][0])
+ ndofs = 5000
+ args = [np.random.randn(ndofs) for i in range(10)]
- assert np.allclose(obj_array_result, ref_result)
-
- # }}}
+ assert_close_to_numpy_in_containers(
+ actx, lambda _np, *_args: _np.stack(_args), args)
def test_actx_concatenate(actx_factory):
@@ -220,39 +221,24 @@ def test_actx_concatenate(actx_factory):
ndofs = 5000
args = [np.random.randn(ndofs) for i in range(10)]
- ref_result = np.concatenate(args)
-
- # {{{ test cl.Arrays
-
- actx_args = [actx.from_numpy(arg) for arg in args]
- actx_result = actx.to_numpy(actx.np.concatenate(actx_args))
-
- assert np.allclose(actx_result, ref_result)
- # }}}
+ assert_close_to_numpy(
+ actx, lambda _np, *_args: _np.concatenate(_args), args)
def test_actx_reshape(actx_factory):
actx = actx_factory()
- numpy_ary = np.random.randn(2, 3)
- actx_ary = actx.from_numpy(numpy_ary)
-
- assert np.allclose(actx.to_numpy(actx.np.reshape(actx_ary, (3, 2))),
- np.reshape(numpy_ary, (3, 2)))
-
- assert np.allclose(actx.to_numpy(actx.np.reshape(actx_ary, (3, -1))),
- np.reshape(numpy_ary, (3, -1)))
-
- assert np.allclose(actx.to_numpy(actx.np.reshape(actx_ary, (6,))),
- np.reshape(numpy_ary, (6,)))
-
- assert np.allclose(actx.to_numpy(actx.np.reshape(actx_ary, -1)),
- np.reshape(numpy_ary, -1))
+ for new_shape in [(3, 2), (3, -1), (6,), (-1,)]:
+ assert_close_to_numpy(
+ actx, lambda _np, *_args: _np.reshape(*_args),
+ (np.random.randn(2, 3), new_shape))
# }}}
+# {{{ arithmetic same as numpy
+
def test_dof_array_arithmetic_same_as_numpy(actx_factory):
actx = actx_factory()
@@ -386,6 +372,10 @@ def test_dof_array_arithmetic_same_as_numpy(actx_factory):
# }}}
+# }}}
+
+
+# {{{ reductions same as numpy
def test_dof_array_reductions_same_as_numpy(actx_factory):
actx = actx_factory()
@@ -399,6 +389,8 @@ def test_dof_array_reductions_same_as_numpy(actx_factory):
assert isinstance(actx_red, Number)
assert np.allclose(np_red, actx_red)
+# }}}
+
# {{{ test array context einsum