take

Returns elements of an array along an axis.

Parameters:

Name	Type	Description	Default
x	SparseArray	Input array.	required
indices	ndarray	Array indices. The array must be one-dimensional and have an integer data type.	required
axis	int	Axis over which to select values. If axis is negative, the function must determine the axis along which to select values by counting from the last dimension. For None, the flattened input array is used. Default: None.	None

Returns:

Name	Type	Description
out	COO	A COO array with requested indices.

Raises:

Type	Description
ValueError	If the input array isn't and can't be converted to COO format.

Source code in sparse/numba_backend/_coo/common.py

def take(x, indices, /, *, axis=None):
    """
    Returns elements of an array along an axis.

    Parameters
    ----------
    x : SparseArray
        Input array.
    indices : ndarray
        Array indices. The array must be one-dimensional and have an integer data type.
    axis : int
        Axis over which to select values. If ``axis`` is negative, the function must
        determine the axis along which to select values by counting from the last dimension.
        For ``None``, the flattened input array is used. Default: ``None``.

    Returns
    -------
    out : COO
        A COO array with requested indices.

    Raises
    ------
    ValueError
        If the input array isn't and can't be converted to COO format.
    """

    x = _validate_coo_input(x)

    if axis is None:
        x = x.flatten()
        return x[indices]

    axis = normalize_axis(axis, x.ndim)
    full_index = (slice(None),) * axis + (indices, ...)
    return x[full_index]