squidpy.im.ImageContainer.generate_equal_crops

ImageContainer.generate_equal_crops(size=None, as_array=False, squeeze=True, **kwargs)

Decompose image into equally sized crops.

Parameters:

• size (int | float | tuple[int | float, int | float] | None) – Size of the crop as (height, width). If a single int, the crop will be a square.

• as_array (str | bool) –

  • If True, yields a dict where keys are layers and values are numpy.ndarray.

  • If a str, yields one numpy.ndarray for the specified layer.

  • If a typing.Sequence, yields a tuple of numpy.ndarray for the specified layers.

  • Otherwise, yields squidpy.im.ImageContainer.

• squeeze (bool) – Remove singleton dimensions from the results if as_array = True.

• kwargs (Any) – Keyword arguments for crop_corner().

Yields:

The crops, whose type depends on as_array.

Return type:

Iterator[ImageContainer] | Iterator[dict[str, ndarray[tuple[Any, ...], dtype[Any]]]]

Notes

Crops going outside out of the image boundary are padded with cval.