unique_counts

Returns the unique elements of an input array x, and the corresponding counts for each unique element in x.

Parameters:

Name	Type	Description	Default
x	COO	Input COO array. It will be flattened if it is not already 1-D.	required

Returns:

Name	Type	Description
out	namedtuple	The result containing: * values - The unique elements of an input array. * counts - The corresponding counts for each unique element.

Raises:

Type	Description
ValueError	If the input array is in a different format than COO.

Examples:

>>> import sparse
>>> x = sparse.COO.from_numpy([1, 0, 2, 1, 2, -3])
>>> sparse.unique_counts(x)
UniqueCountsResult(values=array([-3,  0,  1,  2]), counts=array([1, 1, 2, 2]))

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

def unique_counts(x, /):
    """
    Returns the unique elements of an input array `x`, and the corresponding
    counts for each unique element in `x`.

    Parameters
    ----------
    x : COO
        Input COO array. It will be flattened if it is not already 1-D.

    Returns
    -------
    out : namedtuple
        The result containing:
        * values - The unique elements of an input array.
        * counts - The corresponding counts for each unique element.

    Raises
    ------
    ValueError
        If the input array is in a different format than COO.

    Examples
    --------
    >>> import sparse
    >>> x = sparse.COO.from_numpy([1, 0, 2, 1, 2, -3])
    >>> sparse.unique_counts(x)
    UniqueCountsResult(values=array([-3,  0,  1,  2]), counts=array([1, 1, 2, 2]))
    """

    x = _validate_coo_input(x)

    x = x.flatten()
    values, counts = np.unique(x.data, return_counts=True, equal_nan=False)
    fill_count = x.size - x.nnz
    if fill_count > 0:
        if np.isnan(x.fill_value):
            # Per the Array API spec, NaNs compare as False, so each NaN is distinct.
            values = np.concatenate([values, np.full(fill_count, x.fill_value)])
            counts = np.concatenate([counts, np.ones(fill_count, dtype=counts.dtype)])
        else:
            values = np.concatenate([[x.fill_value], values])
            counts = np.concatenate([[fill_count], counts])
            sorted_indices = np.argsort(values)
            values[sorted_indices] = values.copy()
            counts[sorted_indices] = counts.copy()

    return UniqueCountsResult(values, counts)