aopp_deconv_tool.numpy_helper.slice

module documentation

Helper functions for slice operations on numpy arrays

Function around_centre returns a slice of a in the shape of b about the centre of a Accepts np.ndarray or tuple for each argument
Function from_string Build a tuple of slices from a string representation
Function get_indices Returns the indices of an array slice. Indicies will select the sliced part of an array.
Function iter_indices Iterator that returns the sliced indices of a sliced array.
Function iter_indices_with_slices Iterator that returns the sliced indices of a sliced array, and slices that select only grouped axes
Function squeeze Given a tuple of slices (slice_tuple), if a slice only selects a single index, replace it with an index instead. e.g., [10:100, 13:14, 50:60] -> [10:100, 13, 50:60]
Function unsqueeze Given a tuple of slices and indices (slice_tuple), replace all indices with a slice from i->i+1 e.g., [10:100, 13, 50:60] -> [10:100, 13:14, 50:60]
Variable _lgr Undocumented
def around_centre(big_shape: tuple[int, ...], small_shape: tuple[int, ...]) -> tuple[slice, ...]:

returns a slice of a in the shape of b about the centre of a Accepts np.ndarray or tuple for each argument

def from_string(slice_tuple_str: str) -> tuple[slice, ...]:

Build a tuple of slices from a string representation

def get_indices(a: np.ndarray, slice_tuple: tuple[slice | int, ...] | np.ndarray[int] | None, as_tuple: bool = True) -> np.ndarray | tuple[np.ndarray, ...]:

Returns the indices of an array slice. Indicies will select the sliced part of an array.

a
The array to get the indicies of a slice of
slice_tuple
A tuple of length a.ndim that specifies the slices
as_tuple
If True will return an a.ndim length tuple, otherwise will return a numpy array.
def iter_indices(a: np.ndarray, slice_tuple: tuple[slice | int, ...] | np.ndarray[int] | None = None, group: tuple[int, ...] = tuple(), squeeze=True) -> Iterator[tuple[np.ndarray]]:

Iterator that returns the sliced indices of a sliced array.

a
The array to get the indicies of a slice of
slice_tuple
A tuple of length a.ndim that specifies the slices
group
Axes that are not iterated over (i.e., they are grouped together). E.g. if a.shape=(3,5,7,9) and group=(1,3), then on iteration the indices idx select a slice from a such that a[idx].shape=(5,9).
squeeze
Should non-grouped axes be merged (need one for loop for all of them), or should they remain separate (need a.ndim - len(group) loops).
def iter_indices_with_slices(a: np.ndarray, slice_tuple: tuple[slice | int, ...] | np.ndarray[int] | None = None, group: tuple[int, ...] = tuple(), squeeze=True) -> Iterator[tuple[np.ndarray]]:

Iterator that returns the sliced indices of a sliced array, and slices that select only grouped axes

a
The array to get the indicies of a slice of
slice_tuple
A tuple of length a.ndim that specifies the slices
group
Axes that are not iterated over (i.e., they are grouped together). E.g. if a.shape=(3,5,7,9) and group=(1,3), then on iteration the indices idx select a slice from a such that a[idx].shape=(5,9).
squeeze
Should non-grouped axes be merged (need one for loop for all of them), or should they remain separate (need a.ndim - len(group) loops).
def squeeze(slice_tuple: tuple[slice | int, ...]) -> tuple[slice | int]:

Given a tuple of slices (slice_tuple), if a slice only selects a single index, replace it with an index instead. e.g., [10:100, 13:14, 50:60] -> [10:100, 13, 50:60]

def unsqueeze(slice_tuple: tuple[slice | int, ...]) -> tuple[slice]:

Given a tuple of slices and indices (slice_tuple), replace all indices with a slice from i->i+1 e.g., [10:100, 13, 50:60] -> [10:100, 13:14, 50:60]

_lgr =

Undocumented