def set_src_shape(self, shape, dist_shape=None):
"""
Set the shape of the 'source' array .
Parameters
----------
shape : tuple or int
The shape of the 'source' array.
dist_shape : tuple or None
If not None, the full distributed shape of the source.
Returns
-------
Indexer
Self is returned to allow chaining.
"""
if self._flat_src is None and shape is not None:
self._flat_src = len(shape2tuple(shape)) <= 1
self._src_shape, self._dist_shape, = self._get_shapes(
shape, dist_shape)
if shape is not None:
self._check_bounds()
self._shaped_inst = None
return self
def _get_shapes(self, shape, dist_shape):
if shape is None:
return None, None
shape = shape2tuple(shape)
if self._flat_src:
shape = (np.product(shape, dtype=int), )
if dist_shape is None:
return shape, shape
dist_shape = shape2tuple(dist_shape)
if self._flat_src:
dist_shape = (np.product(dist_shape, dtype=int), )
return shape, dist_shape
def set_src_shape(self, shape, dist_shape=None):
"""
Set the shape of the 'source' array .
Parameters
----------
shape : tuple or int
The shape of the 'source' array.
dist_shape : tuple or None
If not None, the full distributed shape of the source.
Returns
-------
Indexer
Self is returned to allow chaining.
"""
self._check_src_shape(shape2tuple(shape))
super().set_src_shape(shape, dist_shape)
if shape is None:
return self
if self._flat_src:
for i in self._idx_list:
i.set_src_shape(self._src_shape, self._dist_shape)
else:
for i, s, ds in zip(self._idx_list, self._src_shape,
self._dist_shape):
i.set_src_shape(s, ds)
return self
def _get_shapes(self, shape, dist_shape):
if shape is None:
return None, None
shape = shape2tuple(shape)
if self._flat_src:
shape = (shape_to_len(shape), )
if dist_shape is None:
return shape, shape
dist_shape = shape2tuple(dist_shape)
if self._flat_src:
dist_shape = (shape_to_len(dist_shape), )
return shape, dist_shape
def __init__(self,
nvars,
compute_delay=0.001,
compute_partials_delay=0.001,
var_default=1.0,
add_var_kwargs=None,
use_coloring=True,
**kwargs):
super().__init__(**kwargs)
self.nvars = nvars
self.add_var_kwargs = add_var_kwargs
self.compute_delay = compute_delay
self.compute_partials_delay = compute_partials_delay
self.use_coloring = use_coloring
self.inames = []
self.onames = []
self.varshape = () # default to scalar variables
if add_var_kwargs is not None:
if 'shape' in add_var_kwargs:
self.varshape = shape2tuple(add_var_kwargs['shape'])
if 'val' in add_var_kwargs:
v = add_var_kwargs['val']
if isinstance(v, float):
self.varshape = ()
elif isinstance(v, np.ndarray):
self.varshape = v.shape
else:
raise TypeError(
f"ExplicitSleepComp doesn't work with discrete variables."
)
def __init__(self, shape):
"""
Initialize attributes.
Parameters
----------
shape : tuple or int
Shape of the source.
"""
self._shape = shape2tuple(shape)
def _chk_shape_dims(self, flat_src, iname, oname, prefix):
if self._orig_src_shape is None or flat_src or not self._check_dims:
return
if len(self._orig_src_shape) > len(shape2tuple(self.shape)):
issue_warning(
f"connecting source '{oname}' of dimension {len(self._orig_src_shape)} "
f"to '{iname}' using src_indices of dimension {len(self.shape)} without "
"setting `flat_src_indices=True`. The source is currently treated as "
"flat, but this automatic flattening is deprecated and will be removed "
"in a future release. To keep the old behavior, set `flat_src_indices`"
"=True in the connect(), promotes(), or add_input() call.",
category=OMDeprecationWarning,
prefix=prefix)