def _validate_args(self, field_args, param_args, domain, origin) -> None:
"""Validate input arguments to _call_run.
Raises
-------
ValueError
If invalid data or inconsistent options are specified.
TypeError
If an incorrect field or parameter data type is passed.
"""
assert isinstance(field_args, dict) and isinstance(param_args, dict)
# validate domain sizes
domain_ndim = self.domain_info.ndim
if len(domain) != domain_ndim:
raise ValueError(f"Invalid 'domain' value '{domain}'")
try:
domain = Shape(domain)
except:
raise ValueError("Invalid 'domain' value ({})".format(domain))
if not domain > Shape.zeros(domain_ndim):
raise ValueError(f"Compute domain contains zero sizes '{domain}')")
if not domain <= (max_domain := self._get_max_domain(field_args, origin, squeeze=False)):
raise ValueError(
f"Compute domain too large (provided: {domain}, maximum: {max_domain})"
)
def _get_max_domain(self, field_args, origin):
"""Return the maximum domain size possible
Parameters
----------
field_args: `dict`
Mapping from field names to actually passed data arrays.
origin: `{'field_name': [int * ndims]}`
The origin for each field.
Returns
-------
`Shape`: the maximum domain size.
"""
large_val = np.iinfo(np.uintc).max
max_domain = Shape([large_val] * self.domain_info.ndims)
for name, field in field_args.items():
api_mask = self._get_field_mask(name)
if isinstance(field, gt_storage.storage.Storage):
storage_mask = tuple(field.mask)
if storage_mask != api_mask:
raise ValueError(
f"The storage for '{name}' has mask '{storage_mask}', but the API signature expects '{api_mask}'"
)
upper_boundary = self.field_info[
name].boundary.upper_indices.filter_mask(api_mask)
field_domain = Shape(field.shape) - (origin[name] + upper_boundary)
max_domain &= Shape.from_mask(field_domain,
api_mask,
default=large_val)
return max_domain
def _get_max_domain(
self,
field_args: Dict[str, Any],
origin: Dict[str, Tuple[int, ...]],
*,
squeeze: bool = True,
) -> Shape:
"""Return the maximum domain size possible
Parameters
----------
field_args:
Mapping from field names to actually passed data arrays.
origin:
The origin for each field.
squeeze:
Convert non-used domain dimensions to singleton dimensions.
Returns
-------
`Shape`: the maximum domain size.
"""
domain_ndim = self.domain_info.ndim
max_size = sys.maxsize
max_domain = Shape([max_size] * domain_ndim)
for name, field_info in self.field_info.items():
if field_info is not None:
assert field_args.get(
name,
None) is not None, f"Invalid value for '{name}' field."
field = field_args[name]
api_domain_mask = field_info.domain_mask
api_domain_ndim = field_info.domain_ndim
assert (
not isinstance(field, gt_storage.storage.Storage)
or tuple(field.mask)[:domain_ndim] == api_domain_mask
), (f"Storage for '{name}' has domain mask '{field.mask}' but the API signature "
f"expects '[{', '.join(field_info.axes)}]'")
upper_indices = field_info.boundary.upper_indices.filter_mask(
api_domain_mask)
field_origin = Index.from_value(origin[name])
field_domain = tuple(field.shape[i] -
(field_origin[i] + upper_indices[i])
for i in range(api_domain_ndim))
max_domain &= Shape.from_mask(field_domain,
api_domain_mask,
default=max_size)
if squeeze:
return Shape([i if i != max_size else 1 for i in max_domain])
else:
return max_domain
def _get_max_domain(
self,
field_args: Dict[str, Any],
origin: Dict[str, Tuple[int, ...]],
*,
squeeze: bool = True,
) -> Shape:
"""Return the maximum domain size possible.
Parameters
----------
field_args:
Mapping from field names to actually passed data arrays.
origin:
The origin for each field.
squeeze:
Convert non-used domain dimensions to singleton dimensions.
Returns
-------
`Shape`: the maximum domain size.
"""
domain_ndim = self.domain_info.ndim
max_size = sys.maxsize
max_domain = Shape([max_size] * domain_ndim)
for name, field_info in self.field_info.items():
if field_info.access != AccessKind.NONE:
assert field_args.get(
name,
None) is not None, f"Invalid value for '{name}' field."
field = field_args[name]
api_domain_mask = field_info.domain_mask
api_domain_ndim = field_info.domain_ndim
upper_indices = field_info.boundary.upper_indices.filter_mask(
api_domain_mask)
field_origin = Index.from_value(origin[name])
field_domain = tuple(field.shape[i] -
(field_origin[i] + upper_indices[i])
for i in range(api_domain_ndim))
max_domain &= Shape.from_mask(field_domain,
api_domain_mask,
default=max_size)
if squeeze:
return Shape([i if i != max_size else 1 for i in max_domain])
else:
return max_domain
def normalize_shape(
shape: Optional[Sequence[int]], mask: Optional[Sequence[bool]] = None
) -> Optional[Shape]:
check_mask(mask)
if shape is None:
return None
if mask is None:
mask = (True,) * len(shape)
if sum(mask) != len(shape) and len(mask) != len(shape):
raise ValueError(
"len(shape) must be equal to len(mask) or the number of 'True' entries in mask."
)
if not gt_util.is_iterable_of(shape, numbers.Integral):
raise TypeError("shape must be a tuple of ints or pairs of ints.")
if any(o <= 0 for o in shape):
raise ValueError("shape ({}) contains non-positive value.".format(shape))
new_shape = list(shape)
if sum(mask) < len(shape):
new_shape = [int(h) for i, h in enumerate(new_shape) if mask[i]]
return Shape(new_shape)
def _get_max_domain(self, field_args, origin):
"""Return the maximum domain size possible
Parameters
----------
field_args: `dict`
Mapping from field names to actually passed data arrays.
origin: `{'field_name': [int * ndims]}`
The origin for each field.
Returns
-------
`Shape`: the maximum domain size.
"""
max_domain = Shape([np.iinfo(np.uintc).max] * self.domain_info.ndims)
shapes = {name: Shape(field.shape) for name, field in field_args.items()}
for name, shape in shapes.items():
upper_boundary = Index(self.field_info[name].boundary.upper_indices)
max_domain &= shape - (Index(origin[name]) + upper_boundary)
return max_domain
def test_implementation(self, test, parameters_dict):
"""Test computed values for implementations generated for all *backends* and *stencil suites*.
The generated implementations are reused from previous tests by means of a
:class:`utils.ImplementationsDB` instance shared at module scope.
"""
# backend = "debug"
cls = type(self)
implementation_list = test["implementations"]
if not implementation_list:
pytest.skip(
"Cannot perform validation tests, since there are no valid implementations."
)
for implementation in implementation_list:
if not isinstance(implementation, StencilObject):
raise RuntimeError(
"Wrong function got from implementations_db cache!")
fields, exec_info = parameters_dict
# Domain
from gt4py.definitions import Shape
from gt4py.ir.nodes import Index
origin = cls.origin
shapes = {}
for name, field in [(k, v) for k, v in fields.items()
if isinstance(v, np.ndarray)]:
shapes[name] = Shape(field.shape)
max_domain = Shape([sys.maxsize] *
implementation.domain_info.ndims)
for name, shape in shapes.items():
upper_boundary = Index(
[b[1] for b in cls.symbols[name].boundary])
max_domain &= shape - (Index(origin) + upper_boundary)
domain = max_domain
max_boundary = ((0, 0), (0, 0), (0, 0))
for name, info in implementation.field_info.items():
if isinstance(info, gt_definitions.FieldInfo):
max_boundary = tuple(
(max(m[0], abs(b[0])), max(m[1], b[1]))
for m, b in zip(max_boundary, info.boundary))
new_boundary = tuple(
(max(abs(b[0]), abs(mb[0])), max(abs(b[1]), abs(mb[1])))
for b, mb in zip(cls.max_boundary, max_boundary))
shape = None
for name, field in fields.items():
if isinstance(field, np.ndarray):
assert field.shape == (shape if shape is not None else
field.shape)
shape = field.shape
patched_origin = tuple(nb[0] for nb in new_boundary)
patching_origin = tuple(po - o
for po, o in zip(patched_origin, origin))
patched_shape = tuple(nb[0] + nb[1] + d
for nb, d in zip(new_boundary, domain))
patching_slices = [
slice(po, po + s) for po, s in zip(patching_origin, shape)
]
for k, v in implementation.constants.items():
sys.modules[self.__module__].__dict__[k] = v
inputs = {}
for k, f in fields.items():
if isinstance(f, np.ndarray):
patched_f = np.empty(shape=patched_shape)
patched_f[patching_slices] = f
inputs[k] = gt_storage.from_array(
patched_f,
dtype=test["definition"].__annotations__[k],
shape=patched_f.shape,
default_origin=patched_origin,
backend=test["backend"],
)
else:
inputs[k] = f
validation_fields = {
name: np.array(field, copy=True)
for name, field in inputs.items()
}
implementation(**inputs,
origin=patched_origin,
exec_info=exec_info)
domain = exec_info["domain"]
validation_origins = {
name: tuple(
nb[0] - g[0]
for nb, g in zip(new_boundary, cls.symbols[name].boundary))
for name in implementation.field_info.keys()
}
validation_shapes = {
name:
tuple(d + g[0] + g[1]
for d, g in zip(domain, cls.symbols[name].boundary))
for name in implementation.field_info.keys()
}
validation_field_views = {
name: field[tuple(
slice(o, o + s) for o, s in zip(validation_origins[name],
validation_shapes[name]))]
if name in implementation.field_info else field # parameters
for name, field in validation_fields.items()
}
cls.validation(
**validation_field_views,
domain=domain,
origin={
name: tuple(b[0] for b in cls.symbols[name].boundary)
for name in validation_fields
if name in implementation.field_info
},
)
# Test values
for (name,
value), (expected_name,
expected_value) in zip(inputs.items(),
validation_fields.items()):
if isinstance(fields[name], np.ndarray):
domain_slice = [
slice(new_boundary[d][0],
new_boundary[d][0] + domain[d])
for d in range(len(domain))
]
np.testing.assert_allclose(
value.data[domain_slice],
expected_value[domain_slice],
rtol=RTOL,
atol=ATOL,
equal_nan=EQUAL_NAN,
err_msg="Wrong data in output field '{name}'".format(
name=name),
)
def _call_run(self,
field_args,
parameter_args,
domain,
origin,
exec_info=None):
"""Check and preprocess the provided arguments (called by :class:`StencilObject` subclasses).
Note that this function will always try to expand simple parameter values to
complete data structures by repeating the same value as many times as needed.
Parameters
----------
field_args: `dict`
Mapping from field names to actually passed data arrays.
This parameter encapsulates `*args` in the actual stencil subclass
by doing: `{input_name[i]: arg for i, arg in enumerate(args)}`
parameter_args: `dict`
Mapping from parameter names to actually passed parameter values.
This parameter encapsulates `**kwargs` in the actual stencil subclass
by doing: `{name: value for name, value in kwargs.items()}`
domain : `Sequence` of `int`, optional
Shape of the computation domain. If `None`, it will be used the
largest feasible domain according to the provided input fields
and origin values (`None` by default).
origin : `[int * ndims]` or `{'field_name': [int * ndims]}`, optional
If a single offset is passed, it will be used for all fields.
If a `dict` is passed, there could be an entry for each field.
A special key *'_all_'* will represent the value to be used for all
the fields not explicitly defined. If `None` is passed or it is
not possible to assign a value to some field according to the
previous rule, the value will be inferred from the global boundaries
of the field. Note that the function checks if the origin values
are at least equal to the `global_border` attribute of that field,
so a 0-based origin will only be acceptable for fields with
a 0-area support region.
exec_info : `dict`, optional
Dictionary used to store information about the stencil execution.
(`None` by default).
Returns
-------
`None`
Raises
-------
ValueError
If invalid data or inconsistent options are specified.
"""
if exec_info is not None:
exec_info["call_run_start_time"] = time.perf_counter()
used_arg_fields = {
name: field
for name, field in field_args.items()
if name in self.field_info and self.field_info[name] is not None
}
used_arg_params = {
name: param
for name, param in parameter_args.items()
if name in self.parameter_info
and self.parameter_info[name] is not None
}
for name, field_info in self.field_info.items():
if field_info is not None and field_args[name] is None:
raise ValueError(f"Field '{name}' is None.")
for name, parameter_info in self.parameter_info.items():
if parameter_info is not None and parameter_args[name] is None:
raise ValueError(f"Parameter '{name}' is None.")
# assert compatibility of fields with stencil
for name, field in used_arg_fields.items():
if not gt_backend.from_name(
self.backend).storage_info["is_compatible_layout"](field):
raise ValueError(
f"The layout of the field {name} is not compatible with the backend."
)
if not gt_backend.from_name(
self.backend).storage_info["is_compatible_type"](field):
raise ValueError(
f"Field '{name}' has type '{type(field)}', which is not compatible with the '{self.backend}' backend."
)
elif type(field) is np.ndarray:
warnings.warn(
"NumPy ndarray passed as field. This is discouraged and only works with constraints and only for certain backends.",
RuntimeWarning,
)
if not field.dtype == self.field_info[name].dtype:
raise TypeError(
f"The dtype of field '{name}' is '{field.dtype}' instead of '{self.field_info[name].dtype}'"
)
# ToDo: check if mask is correct: need mask info in stencil object.
if isinstance(field, gt_storage.storage.Storage):
if not field.is_stencil_view:
raise ValueError(
f"An incompatible view was passed for field {name} to the stencil. "
)
for name_other, field_other in used_arg_fields.items():
if field_other.mask == field.mask:
if not field_other.shape == field.shape:
raise ValueError(
f"The fields {name} and {name_other} have the same mask but different shapes."
)
# assert compatibility of parameters with stencil
for name, parameter in used_arg_params.items():
if not type(parameter) == self.parameter_info[name].dtype:
raise TypeError(
f"The type of parameter '{name}' is '{type(parameter)}' instead of '{self.parameter_info[name].dtype}'"
)
assert isinstance(field_args, dict) and isinstance(
parameter_args, dict)
# Shapes
shapes = {}
for name, field in used_arg_fields.items():
shapes[name] = Shape(field.shape)
# Origins
if origin is None:
origin = {}
else:
origin = normalize_origin_mapping(origin)
for name, field in used_arg_fields.items():
origin.setdefault(
name,
origin["_all_"] if "_all_" in origin else field.default_origin)
# Domain
max_domain = Shape([sys.maxsize] * self.domain_info.ndims)
for name, shape in shapes.items():
upper_boundary = Index(
self.field_info[name].boundary.upper_indices)
max_domain &= shape - (Index(origin[name]) + upper_boundary)
if domain is None:
domain = max_domain
else:
domain = normalize_domain(domain)
if len(domain) != self.domain_info.ndims:
raise ValueError(f"Invalid 'domain' value '{domain}'")
# check domain+halo vs field size
if not domain > Shape.zeros(self.domain_info.ndims):
raise ValueError(f"Compute domain contains zero sizes '{domain}')")
if not domain <= max_domain:
raise ValueError(
f"Compute domain too large (provided: {domain}, maximum: {max_domain})"
)
for name, field in used_arg_fields.items():
min_origin = self.field_info[name].boundary.lower_indices
if origin[name] < min_origin:
raise ValueError(
f"Origin for field {name} too small. Must be at least {min_origin}, is {origin[name]}"
)
min_shape = tuple(o + d + h for o, d, h in zip(
origin[name], domain,
self.field_info[name].boundary.upper_indices))
if min_shape > field.shape:
raise ValueError(
f"Shape of field {name} is {field.shape} but must be at least {min_shape} for given domain and origin."
)
self.run(**field_args,
**parameter_args,
_domain_=domain,
_origin_=origin,
exec_info=exec_info)
def _run_test_implementation(cls, parameters_dict, implementation): # noqa: C901 # too complex
input_data, exec_info = parameters_dict
origin = cls.origin
max_boundary = Boundary(cls.max_boundary)
field_params = cls.field_params
field_masks = {}
for name, value in input_data.items():
if isinstance(value, np.ndarray):
field_masks[name] = tuple(
ax in field_params[name][0] for ax in CartesianSpace.names
)
data_shape = Shape((sys.maxsize,) * 3)
for name, data in input_data.items():
if isinstance(data, np.ndarray):
data_shape &= Shape(
interpolate_mask(data.shape, field_masks[name], default=sys.maxsize)
)
domain = data_shape - (
Index(max_boundary.lower_indices) + Index(max_boundary.upper_indices)
)
referenced_inputs = {
name: info for name, info in implementation.field_info.items() if info is not None
}
referenced_inputs.update(
{name: info for name, info in implementation.parameter_info.items() if info is not None}
)
# set externals for validation method
for k, v in implementation.constants.items():
sys.modules[cls.__module__].__dict__[k] = v
# copy input data
test_values = {}
validation_values = {}
for name, data in input_data.items():
data = input_data[name]
if name in referenced_inputs:
info = referenced_inputs[name]
if isinstance(info, FieldInfo):
data_dims = field_params[name][1]
if data_dims:
dtype = (data.dtype, data_dims)
shape = data.shape[: -len(data_dims)]
else:
dtype = data.dtype
shape = data.shape
test_values[name] = gt_storage.from_array(
data,
dtype=dtype,
shape=shape,
mask=field_masks[name],
default_origin=origin,
backend=implementation.backend,
)
validation_values[name] = np.array(data)
else:
test_values[name] = data
validation_values[name] = data
else:
test_values[name] = None
validation_values[name] = None
# call implementation
implementation(**test_values, origin=origin, exec_info=exec_info)
assert domain == exec_info["domain"]
# for validation data, data is cropped to actually touched domain, so that origin offseting
# does not have to be implemented for every test suite. This is done based on info
# specified in test suite
cropped_validation_values = {}
for name, data in validation_values.items():
sym = cls.symbols[name]
if data is not None and sym.kind == SymbolKind.FIELD:
field_extent_low = tuple(b[0] for b in sym.boundary)
offset_low = tuple(b[0] - e for b, e in zip(max_boundary, field_extent_low))
field_extent_high = tuple(b[1] for b in sym.boundary)
offset_high = tuple(b[1] - e for b, e in zip(max_boundary, field_extent_high))
validation_slice = filter_mask(
tuple(slice(o, s - h) for o, s, h in zip(offset_low, data_shape, offset_high)),
field_masks[name],
)
data_dims = field_params[name][1]
if data_dims:
validation_slice = tuple([*validation_slice] + [slice(None)] * len(data_dims))
cropped_validation_values[name] = data[validation_slice]
else:
cropped_validation_values[name] = data
cls.validation(
**cropped_validation_values,
domain=domain,
origin={
name: info.boundary.lower_indices
for name, info in implementation.field_info.items()
if info is not None
},
)
# Test values
for name, value in test_values.items():
if isinstance(value, np.ndarray):
expected_value = validation_values[name]
if gt_backend.from_name(value.backend).storage_info["device"] == "gpu":
value.synchronize()
value = value.data.get()
else:
value = value.data
np.testing.assert_allclose(
value,
expected_value,
rtol=RTOL,
atol=ATOL,
equal_nan=EQUAL_NAN,
err_msg="Wrong data in output field '{name}'".format(name=name),
)
def _validate_args(self, used_field_args, used_param_args, domain, origin):
"""Validate input arguments to _call_run.
Raises
-------
ValueError
If invalid data or inconsistent options are specified.
TypeError
If an incorrect field or parameter data type is passed.
"""
# assert compatibility of fields with stencil
for name, field in used_field_args.items():
if not gt_backend.from_name(
self.backend).storage_info["is_compatible_layout"](field):
raise ValueError(
f"The layout of the field {name} is not compatible with the backend."
)
if not gt_backend.from_name(
self.backend).storage_info["is_compatible_type"](field):
raise ValueError(
f"Field '{name}' has type '{type(field)}', which is not compatible with the '{self.backend}' backend."
)
elif type(field) is np.ndarray:
warnings.warn(
"NumPy ndarray passed as field. This is discouraged and only works with constraints and only for certain backends.",
RuntimeWarning,
)
field_dtype = self.field_info[name].dtype
if not field.dtype == field_dtype:
raise TypeError(
f"The dtype of field '{name}' is '{field.dtype}' instead of '{field_dtype}'"
)
if isinstance(field, gt_storage.storage.Storage):
field_mask = self._get_field_mask(name)
storage_mask = tuple(field.mask)
if storage_mask != field_mask:
raise ValueError(
f"The storage for '{name}' has mask '{storage_mask}', but the API signature expects '{field_mask}'"
)
if not field.is_stencil_view:
raise ValueError(
f"An incompatible view was passed for field {name} to the stencil. "
)
# assert compatibility of parameters with stencil
for name, parameter in used_param_args.items():
if not type(parameter) == self.parameter_info[name].dtype:
raise TypeError(
f"The type of parameter '{name}' is '{type(parameter)}' instead of '{self.parameter_info[name].dtype}'"
)
assert isinstance(used_field_args, dict) and isinstance(
used_param_args, dict)
if len(domain) != self.domain_info.ndims:
raise ValueError(f"Invalid 'domain' value '{domain}'")
# check domain+halo vs field size
if not domain > Shape.zeros(self.domain_info.ndims):
raise ValueError(f"Compute domain contains zero sizes '{domain}')")
max_domain = self._get_max_domain(used_field_args, origin)
if not domain <= max_domain:
raise ValueError(
f"Compute domain too large (provided: {domain}, maximum: {max_domain})"
)
for name, field in used_field_args.items():
field_mask = self._get_field_mask(name)
min_origin = self.field_info[
name].boundary.lower_indices.filter_mask(field_mask)
restricted_domain = domain.filter_mask(field_mask)
upper_indices = self.field_info[
name].boundary.upper_indices.filter_mask(field_mask)
if origin[name] < min_origin:
raise ValueError(
f"Origin for field {name} too small. Must be at least {min_origin}, is {origin[name]}"
)
min_shape = tuple(o + d + h for o, d, h in zip(
origin[name], restricted_domain, upper_indices))
if min_shape > field.shape:
raise ValueError(
f"Shape of field {name} is {field.shape} but must be at least {min_shape} for given domain and origin."
)
def _run_test_implementation(self, parameters_dict, implementation):
cls = type(self)
fields, exec_info = parameters_dict
# Domain
from gt4py.definitions import Shape
from gt4py.ir.nodes import Index
origin = cls.origin
shapes = {}
for name, field in [(k, v) for k, v in fields.items() if isinstance(v, np.ndarray)]:
shapes[name] = Shape(field.shape)
max_domain = Shape([sys.maxsize] * implementation.domain_info.ndims)
for name, shape in shapes.items():
upper_boundary = Index([b[1] for b in cls.symbols[name].boundary])
max_domain &= shape - (Index(origin) + upper_boundary)
domain = max_domain
max_boundary = ((0, 0), (0, 0), (0, 0))
for info in implementation.field_info.values():
if isinstance(info, gt_definitions.FieldInfo):
max_boundary = tuple(
(max(m[0], abs(b[0])), max(m[1], b[1]))
for m, b in zip(max_boundary, info.boundary)
)
new_boundary = tuple(
(max(abs(b[0]), abs(mb[0])), max(abs(b[1]), abs(mb[1])))
for b, mb in zip(cls.max_boundary, max_boundary)
)
shape = None
for field in fields.values():
if isinstance(field, np.ndarray):
assert field.shape == (shape if shape is not None else field.shape)
shape = field.shape
patched_origin = tuple(nb[0] for nb in new_boundary)
patching_origin = tuple(po - o for po, o in zip(patched_origin, origin))
patched_shape = tuple(nb[0] + nb[1] + d for nb, d in zip(new_boundary, domain))
patching_slices = [slice(po, po + s) for po, s in zip(patching_origin, shape)]
for k, v in implementation.constants.items():
sys.modules[self.__module__].__dict__[k] = v
inputs = {}
for k, f in fields.items():
if isinstance(f, np.ndarray):
patched_f = np.empty(shape=patched_shape)
patched_f[patching_slices] = f
inputs[k] = gt_storage.from_array(
patched_f,
dtype=f.dtype,
shape=patched_f.shape,
default_origin=patched_origin,
backend=implementation.backend,
)
else:
inputs[k] = f
# remove unused input parameters
inputs = {key: value for key, value in inputs.items() if value is not None}
validation_fields = {name: np.array(field, copy=True) for name, field in inputs.items()}
implementation(**inputs, origin=patched_origin, exec_info=exec_info)
domain = exec_info["domain"]
validation_origins = {
name: tuple(nb[0] - g[0] for nb, g in zip(new_boundary, cls.symbols[name].boundary))
for name in inputs
if name in implementation.field_info
}
validation_shapes = {
name: tuple(d + g[0] + g[1] for d, g in zip(domain, cls.symbols[name].boundary))
for name in inputs
if name in implementation.field_info
}
validation_field_views = {
name: field[
tuple(
slice(o, o + s)
for o, s in zip(validation_origins[name], validation_shapes[name])
)
]
if name in implementation.field_info
else field # parameters
for name, field in validation_fields.items()
}
cls.validation(
**validation_field_views,
domain=domain,
origin={
name: tuple(b[0] for b in cls.symbols[name].boundary)
for name in validation_fields
if name in implementation.field_info
},
)
# Test values
for (name, value), expected_value in zip(inputs.items(), validation_fields.values()):
if isinstance(fields[name], np.ndarray):
domain_slice = [
slice(new_boundary[d][0], new_boundary[d][0] + domain[d])
for d in range(len(domain))
]
if gt_backend.from_name(value.backend).storage_info["device"] == "gpu":
value.synchronize()
value = value.data.get()
else:
value = value.data
np.testing.assert_allclose(
value[domain_slice],
expected_value[domain_slice],
rtol=RTOL,
atol=ATOL,
equal_nan=EQUAL_NAN,
err_msg="Wrong data in output field '{name}'".format(name=name),
)
def _run_test_implementation(self, parameters_dict, implementation):
cls = type(self)
input_data, exec_info = parameters_dict
origin = cls.origin
max_boundary = Boundary(cls.max_boundary)
shape_iter = (Shape(v.shape) for v in input_data.values()
if isinstance(v, np.ndarray))
shape = next(shape_iter)
assert all(shape == sh for sh in shape_iter)
domain = shape - (Index(max_boundary.lower_indices) +
Index(max_boundary.upper_indices))
referenced_inputs = {
name: info
for name, info in implementation.field_info.items()
if info is not None
}
referenced_inputs.update({
name: info
for name, info in implementation.parameter_info.items()
if info is not None
})
# set externals for validation method
for k, v in implementation.constants.items():
sys.modules[self.__module__].__dict__[k] = v
# copy input data
inputs = {}
validation_inputs = {}
for name, data in input_data.items():
data = input_data[name]
if name in referenced_inputs:
info = referenced_inputs[name]
if isinstance(info, FieldInfo):
inputs[name] = gt_storage.from_array(
data,
dtype=data.dtype,
shape=shape,
default_origin=origin,
backend=implementation.backend,
)
validation_inputs[name] = np.array(data)
else:
inputs[name] = data
validation_inputs[name] = data
else:
inputs[name] = None
validation_inputs[name] = None
# call implementation
implementation(**inputs, origin=origin, exec_info=exec_info)
assert domain == exec_info["domain"]
# for validation data, data is cropped to actually touched domain, so that origin offseting
# does not have to be implemented for every test suite. This is done based on info
# specified in test suite
cropped_validation_inputs = {}
for name, data in validation_inputs.items():
sym = cls.symbols[name]
if data is not None and sym.kind == SymbolKind.FIELD:
field_extent_low = tuple(b[0] for b in sym.boundary)
offset_low = tuple(
b[0] - e for b, e in zip(max_boundary, field_extent_low))
field_extent_high = tuple(b[1] for b in sym.boundary)
offset_high = tuple(
b[1] - e for b, e in zip(max_boundary, field_extent_high))
validation_slice = tuple(
slice(o, s - h)
for o, s, h in zip(offset_low, shape, offset_high))
cropped_validation_inputs[name] = data[validation_slice]
else:
cropped_validation_inputs[name] = data
cls.validation(
**cropped_validation_inputs,
domain=domain,
origin={
name: info.boundary.lower_indices
for name, info in implementation.field_info.items()
if info is not None
},
)
# Test values
for name, value in inputs.items():
if isinstance(value, np.ndarray):
expected_value = validation_inputs[name]
if gt_backend.from_name(
value.backend).storage_info["device"] == "gpu":
value.synchronize()
value = value.data.get()
else:
value = value.data
np.testing.assert_allclose(
value,
expected_value,
rtol=RTOL,
atol=ATOL,
equal_nan=EQUAL_NAN,
err_msg="Wrong data in output field '{name}'".format(
name=name),
)