Ctrl+K

unique_all#

ivy.unique_all(x, /, *, axis=None, by_value=True)[source]#

Return the unique elements of an input array x, the first occurring indices for each unique element in x, the indices from the set of unique elements that reconstruct x, and the corresponding counts for each unique element in x.

Data-dependent output shape

The shapes of two of the output arrays for this function depend on the data values in the input array; hence, array libraries which build computation graphs (e.g., JAX, Dask, etc.) may find this function difficult to implement without knowing array values. Accordingly, such libraries may choose to omit this function. See data-dependent-output-shapes section for more details.

Note

Uniqueness should be determined based on value equality (i.e., x_i == x_j). For input arrays having floating-point data types, value-based equality implies the following behavior.

  • As nan values compare as False, nan values should be considered distinct.

  • As -0 and +0 compare as True, signed zeros should not be considered distinct, and the corresponding unique element will be implementation-dependent (e.g., an implementation could choose to return -0 if -0 occurs before +0).

As signed zeros are not distinct, using inverse_indices to reconstruct the input array is not guaranteed to return an array having the exact same values.

Each nan value should have a count of one, while the counts for signed zeros should be aggregated as a single count.

Parameters:

  • x (Union[Array, NativeArray]) – input array.

  • axis (Optional[int], default: None) – the axis to apply unique on. If None, the unique elements of the flattened x are returned.

  • by_value (bool, default: True) – If False, the unique elements will be sorted in the same order that they occur in ‘’x’’. Otherwise, they will be sorted by value.

Return type:

Tuple[Union[Array, NativeArray], Union[Array, NativeArray], Union[Array, NativeArray], Union[Array, NativeArray]]

Returns:

ret – a namedtuple (values, indices, inverse_indices, counts) whose - first element must have the field name values and must be an array

containing the unique elements of x. The array must have the same data type as x.

  • second element must have the field name indices and must be an array containing the indices (first occurrences) of x that result in values. The array must have the same length as values and must have the default array index data type.

  • third element must have the field name inverse_indices and must be an array containing the indices of values that reconstruct x. The array must have the same length as the axis dimension of x and must have the default array index data type.

  • fourth element must have the field name counts and must be an array containing the number of times each unique element occurs in x. The returned array must have the same length as values and must have the default array index data type.

This function conforms to the Array API Standard. This docstring is an extension of the docstring in the standard.

Both the description and the type hints above assumes an array input for simplicity, but this function is nestable, and therefore also accepts ivy.Container instances in place of any of the arguments.

Examples

With ivy.Array input:

>>> x = ivy.randint(0, 10, shape=(2, 2), seed=0)
>>> z = ivy.unique_all(x)
>>> print(z)
Results(values=ivy.array([1, 2, 5, 9]),
        indices=ivy.array([3, 2, 1, 0]),
        inverse_indices=ivy.array([[3, 2], [1, 0]]),
       counts=ivy.array([1, 1, 1, 1]))

>>> x = ivy.array([[ 2.1141,  0.8101,  0.9298,  0.8460],
...                       [-1.2119, -0.3519, -0.6252,  0.4033],
...                       [ 0.7443,  0.2577, -0.3707, -0.0545],
...                       [-0.3238,  0.5944,  0.0775, -0.4327]])
>>> x[range(4), range(4)] = ivy.nan #Introduce NaN values
>>> z = ivy.unique_all(x)
>>> print(z)
Results(values=ivy.array([-1.2119    , -0.62519997, -0.3238    , -0.0545    ,
    0.0775    ,    0.2577    ,  0.40329999,  0.59439999,  0.74430001,  0.81010002,
    0.84600002,  0.92979997,         nan,         nan,         nan,         nan]),
    indices=ivy.array([ 4,  6, 12, 11, 14,  9,  7, 13,  8,  1,  3,  2,  0,  5,
                        10, 15]),
    inverse_indices=ivy.array([[12,  9, 11, 10],
                               [ 0, 12,  1,  6],
                               [ 8,  5, 12,  3],
                               [ 2,  7,  4, 12]]),
   counts=ivy.array([1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]))
Array.unique_all(self, /, *, axis=None, by_value=True)[source]#

ivy.Array instance method variant of ivy.unique_all. This method simply wraps the function, and so the docstring for ivy.unique_all also applies to this method with minimal changes.

Parameters:

  • self (Array) – input array.

  • axis (Optional[int], default: None) – the axis to apply unique on. If None, the unique elements of the flattened x are returned.

  • by_value (bool, default: True) – If False, the unique elements will be sorted in the same order that they occur in ‘’x’’. Otherwise, they will be sorted by value.

Return type:

Tuple[Array, Array, Array, Array]

Returns:

ret – a namedtuple (values, indices, inverse_indices, counts). The details can be found in the docstring for ivy.unique_all.

Examples

>>> x = ivy.randint(0, 10, shape=(2, 2), seed=0)
>>> z = x.unique_all()
>>> print(z)
Results(values=ivy.array([1, 2, 5, 9]),
        indices=ivy.array([3, 2, 1, 0]),
        inverse_indices=ivy.array([[3, 2], [1, 0]]),
       counts=ivy.array([1, 1, 1, 1]))
Container.unique_all(self, /, *, axis=None, by_value=True, key_chains=None, to_apply=True, prune_unapplied=False, map_sequences=False)[source]#

ivy.Container instance method variant of ivy.unique_all. This method simply wraps the function, and so the docstring for ivy.unique_all also applies to this method with minimal changes.

Parameters:

  • self (Container) – input container.

  • axis (Optional[Union[int, Container]], default: None) – the axis to apply unique on. If None, the unique elements of the flattened x are returned.

  • by_value (Union[bool, Container], default: True) – If False, the unique elements will be sorted in the same order that they occur in ‘’x’’. Otherwise, they will be sorted by value.

  • key_chains (Optional[Union[List[str], Dict[str, str], Container]], default: None) – The key-chains to apply or not apply the method to. Default is None.

  • to_apply (Union[bool, Container], default: True) – If True, the method will be applied to key_chains, otherwise key_chains will be skipped. Default is True.

  • prune_unapplied (Union[bool, Container], default: False) – Whether to prune key_chains for which the function was not applied. Default is False.

  • map_sequences (Union[bool, Container], default: False) – Whether to also map method to sequences (lists, tuples). Default is False.

Return type:

Container

Returns:

ret – A container of namedtuples (values, indices, inverse_indices, counts). The details of each entry can be found in the docstring for ivy.unique_all.

Examples

>>> x = ivy.Container(a=ivy.array([0., 1., 3. , 2. , 1. , 0.]),
...                   b=ivy.array([1,2,1,3,4,1,3]))
>>> y = x.unique_all()
>>> print(y)
[{
    a: ivy.array([0., 1., 2., 3.]),
    b: ivy.array([1, 2, 3, 4])
}, {
    a: ivy.array([0, 1, 3, 2]),
    b: ivy.array([0, 1, 3, 4])
}, {
    a: ivy.array([0, 1, 3, 2, 1, 0]),
    b: ivy.array([0, 1, 0, 2, 3, 0, 2])
}, {
    a: ivy.array([2, 2, 1, 1]),
    b: ivy.array([3, 1, 2, 1])
}]
On this page