API References

OpTree: Optimized PyTree Utilities.

optree.MAX_RECURSION_DEPTH: int = 1000

Maximum recursion depth for pytree traversal.

This limit prevents infinite recursion from causing an overflow of the C stack and crashing Python.

optree.NONE_IS_NODE: bool = False: Literal constant that treats None as a pytree non-leaf node.

optree.NONE_IS_LEAF: bool = True: Literal constant that treats None as a pytree leaf node.

optree.tree_flatten(tree, /, is_leaf=None, *, none_is_leaf=False, namespace='')[source]

Flatten a pytree.

The flattening order (i.e., the order of elements in the output list) is deterministic, corresponding to a left-to-right depth-first tree traversal.

>>> tree = {'b': (2, [3, 4]), 'a': 1, 'c': None, 'd': 5}
>>> tree_flatten(tree)
(
    [1, 2, 3, 4, 5],
    PyTreeSpec({'a': *, 'b': (*, [*, *]), 'c': None, 'd': *})
)
>>> tree_flatten(tree, none_is_leaf=True)
(
    [1, 2, 3, 4, None, 5],
    PyTreeSpec({'a': *, 'b': (*, [*, *]), 'c': *, 'd': *}, NoneIsLeaf)
)
>>> tree_flatten(1)
([1], PyTreeSpec(*))
>>> tree_flatten(None)
([], PyTreeSpec(None))
>>> tree_flatten(None, none_is_leaf=True)
([None], PyTreeSpec(*, NoneIsLeaf))

For unordered dictionaries, dict and collections.defaultdict, the order is dependent on the sorted keys in the dictionary. Please use collections.OrderedDict if you want to keep the keys in the insertion order.

>>> from collections import OrderedDict
>>> tree = OrderedDict([('b', (2, [3, 4])), ('a', 1), ('c', None), ('d', 5)])
>>> tree_flatten(tree)
(
    [2, 3, 4, 1, 5],
    PyTreeSpec(OrderedDict({'b': (*, [*, *]), 'a': *, 'c': None, 'd': *}))
)
>>> tree_flatten(tree, none_is_leaf=True)
(
    [2, 3, 4, 1, None, 5],
    PyTreeSpec(OrderedDict({'b': (*, [*, *]), 'a': *, 'c': *, 'd': *}), NoneIsLeaf)
)

Parameters:

tree (pytree) – A pytree to flatten.
is_leaf (callable, optional) – An optionally specified function that will be called at each flattening step. It should return a boolean, with True stopping the traversal and the whole subtree being treated as a leaf, and False indicating the flattening should traverse the current object.
none_is_leaf (bool, optional) – Whether to treat None as a leaf. If False, None is a non-leaf node with arity 0. Thus None is contained in the treespec rather than in the leaves list. (default: False)
namespace (str, optional) – The registry namespace used for custom pytree node types. (default: '', i.e., the global namespace)

Return type:

tuple[list[TypeVar(T)], PyTreeSpec]

Returns:

A pair (leaves, treespec) where the first element is a list of leaf values and the second element is a treespec representing the structure of the pytree.

optree.tree_flatten_with_path(tree, /, is_leaf=None, *, none_is_leaf=False, namespace='')[source]

Flatten a pytree and additionally record the paths.

The flattening order (i.e., the order of elements in the output list) is deterministic, corresponding to a left-to-right depth-first tree traversal.

>>> tree = {'b': (2, [3, 4]), 'a': 1, 'c': None, 'd': 5}
>>> tree_flatten_with_path(tree)
(
    [('a',), ('b', 0), ('b', 1, 0), ('b', 1, 1), ('d',)],
    [1, 2, 3, 4, 5],
    PyTreeSpec({'a': *, 'b': (*, [*, *]), 'c': None, 'd': *})
)
>>> tree_flatten_with_path(tree, none_is_leaf=True)
(
    [('a',), ('b', 0), ('b', 1, 0), ('b', 1, 1), ('c',), ('d',)],
    [1, 2, 3, 4, None, 5],
    PyTreeSpec({'a': *, 'b': (*, [*, *]), 'c': *, 'd': *}, NoneIsLeaf)
)
>>> tree_flatten_with_path(1)
([()], [1], PyTreeSpec(*))
>>> tree_flatten_with_path(None)
([], [], PyTreeSpec(None))
>>> tree_flatten_with_path(None, none_is_leaf=True)
([()], [None], PyTreeSpec(*, NoneIsLeaf))

For unordered dictionaries, dict and collections.defaultdict, the order is dependent on the sorted keys in the dictionary. Please use collections.OrderedDict if you want to keep the keys in the insertion order.

>>> from collections import OrderedDict
>>> tree = OrderedDict([('b', (2, [3, 4])), ('a', 1), ('c', None), ('d', 5)])
>>> tree_flatten_with_path(tree)
(
    [('b', 0), ('b', 1, 0), ('b', 1, 1), ('a',), ('d',)],
    [2, 3, 4, 1, 5],
    PyTreeSpec(OrderedDict({'b': (*, [*, *]), 'a': *, 'c': None, 'd': *}))
)
>>> tree_flatten_with_path(tree, none_is_leaf=True)
(
    [('b', 0), ('b', 1, 0), ('b', 1, 1), ('a',), ('c',), ('d',)],
    [2, 3, 4, 1, None, 5],
    PyTreeSpec(OrderedDict({'b': (*, [*, *]), 'a': *, 'c': *, 'd': *}), NoneIsLeaf)
)

Parameters:

tree (pytree) – A pytree to flatten.
is_leaf (callable, optional) – An optionally specified function that will be called at each flattening step. It should return a boolean, with True stopping the traversal and the whole subtree being treated as a leaf, and False indicating the flattening should traverse the current object.
none_is_leaf (bool, optional) – Whether to treat None as a leaf. If False, None is a non-leaf node with arity 0. Thus None is contained in the treespec rather than in the leaves list. (default: False)
namespace (str, optional) – The registry namespace used for custom pytree node types. (default: '', i.e., the global namespace)

Return type:

tuple[list[tuple[Any, ...]], list[TypeVar(T)], PyTreeSpec]

Returns:

A triple (paths, leaves, treespec). The first element is a list of the paths to the leaf values, while each path is a tuple of the index or keys. The second element is a list of leaf values and the last element is a treespec representing the structure of the pytree.

optree.tree_flatten_with_accessor(tree, /, is_leaf=None, *, none_is_leaf=False, namespace='')[source]

Flatten a pytree and additionally record the accessors.

The flattening order (i.e., the order of elements in the output list) is deterministic, corresponding to a left-to-right depth-first tree traversal.

>>> tree = {'b': (2, [3, 4]), 'a': 1, 'c': None, 'd': 5}
>>> tree_flatten_with_accessor(tree)
(
    [
        PyTreeAccessor(*['a'], (MappingEntry(key='a', type=<class 'dict'>),)),
        PyTreeAccessor(*['b'][0], (MappingEntry(key='b', type=<class 'dict'>), SequenceEntry(index=0, type=<class 'tuple'>))),
        PyTreeAccessor(*['b'][1][0], (MappingEntry(key='b', type=<class 'dict'>), SequenceEntry(index=1, type=<class 'tuple'>), SequenceEntry(index=0, type=<class 'list'>))),
        PyTreeAccessor(*['b'][1][1], (MappingEntry(key='b', type=<class 'dict'>), SequenceEntry(index=1, type=<class 'tuple'>), SequenceEntry(index=1, type=<class 'list'>))),
        PyTreeAccessor(*['d'], (MappingEntry(key='d', type=<class 'dict'>),))
    ],
    [1, 2, 3, 4, 5],
    PyTreeSpec({'a': *, 'b': (*, [*, *]), 'c': None, 'd': *})
)
>>> tree_flatten_with_accessor(tree, none_is_leaf=True)
(
    [
        PyTreeAccessor(*['a'], (MappingEntry(key='a', type=<class 'dict'>),)),
        PyTreeAccessor(*['b'][0], (MappingEntry(key='b', type=<class 'dict'>), SequenceEntry(index=0, type=<class 'tuple'>))),
        PyTreeAccessor(*['b'][1][0], (MappingEntry(key='b', type=<class 'dict'>), SequenceEntry(index=1, type=<class 'tuple'>), SequenceEntry(index=0, type=<class 'list'>))),
        PyTreeAccessor(*['b'][1][1], (MappingEntry(key='b', type=<class 'dict'>), SequenceEntry(index=1, type=<class 'tuple'>), SequenceEntry(index=1, type=<class 'list'>))),
        PyTreeAccessor(*['c'], (MappingEntry(key='c', type=<class 'dict'>),)),
        PyTreeAccessor(*['d'], (MappingEntry(key='d', type=<class 'dict'>),))
    ],
    [1, 2, 3, 4, None, 5],
    PyTreeSpec({'a': *, 'b': (*, [*, *]), 'c': *, 'd': *}, NoneIsLeaf)
)
>>> tree_flatten_with_accessor(1)
([PyTreeAccessor(*, ())], [1], PyTreeSpec(*))
>>> tree_flatten_with_accessor(None)
([], [], PyTreeSpec(None))
>>> tree_flatten_with_accessor(None, none_is_leaf=True)
([PyTreeAccessor(*, ())], [None], PyTreeSpec(*, NoneIsLeaf))

For unordered dictionaries, dict and collections.defaultdict, the order is dependent on the sorted keys in the dictionary. Please use collections.OrderedDict if you want to keep the keys in the insertion order.

>>> from collections import OrderedDict
>>> tree = OrderedDict([('b', (2, [3, 4])), ('a', 1), ('c', None), ('d', 5)])
>>> tree_flatten_with_accessor(tree)
(
    [
        PyTreeAccessor(*['b'][0], (MappingEntry(key='b', type=<class 'collections.OrderedDict'>), SequenceEntry(index=0, type=<class 'tuple'>))),
        PyTreeAccessor(*['b'][1][0], (MappingEntry(key='b', type=<class 'collections.OrderedDict'>), SequenceEntry(index=1, type=<class 'tuple'>), SequenceEntry(index=0, type=<class 'list'>))),
        PyTreeAccessor(*['b'][1][1], (MappingEntry(key='b', type=<class 'collections.OrderedDict'>), SequenceEntry(index=1, type=<class 'tuple'>), SequenceEntry(index=1, type=<class 'list'>))),
        PyTreeAccessor(*['a'], (MappingEntry(key='a', type=<class 'collections.OrderedDict'>),)),
        PyTreeAccessor(*['d'], (MappingEntry(key='d', type=<class 'collections.OrderedDict'>),))
    ],
    [2, 3, 4, 1, 5],
    PyTreeSpec(OrderedDict({'b': (*, [*, *]), 'a': *, 'c': None, 'd': *}))
)
>>> tree_flatten_with_accessor(tree, none_is_leaf=True)
(
    [
        PyTreeAccessor(*['b'][0], (MappingEntry(key='b', type=<class 'collections.OrderedDict'>), SequenceEntry(index=0, type=<class 'tuple'>))),
        PyTreeAccessor(*['b'][1][0], (MappingEntry(key='b', type=<class 'collections.OrderedDict'>), SequenceEntry(index=1, type=<class 'tuple'>), SequenceEntry(index=0, type=<class 'list'>))),
        PyTreeAccessor(*['b'][1][1], (MappingEntry(key='b', type=<class 'collections.OrderedDict'>), SequenceEntry(index=1, type=<class 'tuple'>), SequenceEntry(index=1, type=<class 'list'>))),
        PyTreeAccessor(*['a'], (MappingEntry(key='a', type=<class 'collections.OrderedDict'>),)),
        PyTreeAccessor(*['c'], (MappingEntry(key='c', type=<class 'collections.OrderedDict'>),)),
        PyTreeAccessor(*['d'], (MappingEntry(key='d', type=<class 'collections.OrderedDict'>),))
    ],
    [2, 3, 4, 1, None, 5],
    PyTreeSpec(OrderedDict({'b': (*, [*, *]), 'a': *, 'c': *, 'd': *}), NoneIsLeaf)
)

Parameters:

tree (pytree) – A pytree to flatten.
is_leaf (callable, optional) – An optionally specified function that will be called at each flattening step. It should return a boolean, with True stopping the traversal and the whole subtree being treated as a leaf, and False indicating the flattening should traverse the current object.
none_is_leaf (bool, optional) – Whether to treat None as a leaf. If False, None is a non-leaf node with arity 0. Thus None is contained in the treespec rather than in the leaves list. (default: False)
namespace (str, optional) – The registry namespace used for custom pytree node types. (default: '', i.e., the global namespace)

Return type:

tuple[list[PyTreeAccessor], list[TypeVar(T)], PyTreeSpec]

Returns:

A triple (accessors, leaves, treespec). The first element is a list of accessors to the leaf values. The second element is a list of leaf values and the last element is a treespec representing the structure of the pytree.

optree.tree_unflatten(treespec, leaves)[source]

Reconstruct a pytree from the treespec and the leaves.

The inverse of tree_flatten().

>>> tree = {'b': (2, [3, 4]), 'a': 1, 'c': None, 'd': 5}
>>> leaves, treespec = tree_flatten(tree)
>>> tree == tree_unflatten(treespec, leaves)
True

Parameters:

treespec (PyTreeSpec) – The treespec to reconstruct.
leaves (iterable) – The list of leaves to use for reconstruction. The list must match the number of leaves of the treespec.

Return type:

PyTree [TypeVar(T)]

Returns:

The reconstructed pytree, containing the leaves placed in the structure described by treespec.

optree.tree_iter(tree, /, is_leaf=None, *, none_is_leaf=False, namespace='')[source]

Get an iterator over the leaves of a pytree.

See also tree_flatten() and tree_leaves().

>>> tree = {'b': (2, [3, 4]), 'a': 1, 'c': None, 'd': 5}
>>> list(tree_iter(tree))
[1, 2, 3, 4, 5]
>>> list(tree_iter(tree, none_is_leaf=True))
[1, 2, 3, 4, None, 5]
>>> list(tree_iter(1))
[1]
>>> list(tree_iter(None))
[]
>>> list(tree_iter(None, none_is_leaf=True))
[None]

Parameters:

tree (pytree) – A pytree to iterate over.
is_leaf (callable, optional) – An optionally specified function that will be called at each flattening step. It should return a boolean, with True stopping the traversal and the whole subtree being treated as a leaf, and False indicating the flattening should traverse the current object.
none_is_leaf (bool, optional) – Whether to treat None as a leaf. If False, None is a non-leaf node with arity 0. Thus None is contained in the treespec rather than in the leaves list. (default: False)
namespace (str, optional) – The registry namespace used for custom pytree node types. (default: '', i.e., the global namespace)

Return type:

Iterable[TypeVar(T)]

Returns:

An iterator over the leaf values.

optree.tree_leaves(tree, /, is_leaf=None, *, none_is_leaf=False, namespace='')[source]

Get the leaves of a pytree.

See also tree_flatten() and tree_iter().

>>> tree = {'b': (2, [3, 4]), 'a': 1, 'c': None, 'd': 5}
>>> tree_leaves(tree)
[1, 2, 3, 4, 5]
>>> tree_leaves(tree, none_is_leaf=True)
[1, 2, 3, 4, None, 5]
>>> tree_leaves(1)
[1]
>>> tree_leaves(None)
[]
>>> tree_leaves(None, none_is_leaf=True)
[None]

Parameters:

tree (pytree) – A pytree to flatten.
is_leaf (callable, optional) – An optionally specified function that will be called at each flattening step. It should return a boolean, with True stopping the traversal and the whole subtree being treated as a leaf, and False indicating the flattening should traverse the current object.
none_is_leaf (bool, optional) – Whether to treat None as a leaf. If False, None is a non-leaf node with arity 0. Thus None is contained in the treespec rather than in the leaves list. (default: False)
namespace (str, optional) – The registry namespace used for custom pytree node types. (default: '', i.e., the global namespace)

Return type:

list[TypeVar(T)]

Returns:

A list of leaf values.

optree.tree_structure(tree, /, is_leaf=None, *, none_is_leaf=False, namespace='')[source]

Get the treespec for a pytree.

See also tree_transpose_map().

>>> left, right = tree_partition(lambda x: x > 10, {'x': 7, 'y': (42, 64)})
>>> left
{'x': None, 'y': (42, 64)}
>>> right
{'x': 7, 'y': (None, None)}

Instead of None, one can also use a different sentinel value:

>>> sentinel = object()
>>> left, right = tree_partition(lambda x: x > 10, {'x': 7, 'y': (42, 64)}, fillvalue=sentinel)
>>> left
{'x': <object object at ...>, 'y': (42, 64)}
>>> right
{'x': 7, 'y': (<object object at ...>, <object object at ...>)}

Parameters:

predicate (callable) – A function that takes a leaf value as argument, and splits/partitions it into the left or right tree based on the predicates return value.
tree (pytree) – A pytree to be split, with each leaf providing the first positional argument to function predicate.
is_leaf (callable, optional) – An optionally specified function that will be called at each flattening step. It should return a boolean, with True stopping the traversal and the whole subtree being treated as a leaf, and False indicating the flattening should traverse the current object.
fillvalue (object, optional) – A sentinel value to retain the tree structure. (default: None)
none_is_leaf (bool, optional) – Whether to treat None as a leaf. If False, None is a non-leaf node with arity 0. Thus None is contained in the treespec rather than in the leaves list and None will be remain in the result pytree. (default: False)
namespace (str, optional) – The registry namespace used for custom pytree node types. (default: '', i.e., the global namespace)

Returns:

Two pytrees with the same structure as tree but with orthogonal leaves based on the predicate function. The first pytree contains all leaves where predicate evaluates to True, the second for False. The removed nodes in both trees are filled with fillvalue to keep the original tree structure.

optree.tree_transpose(outer_treespec, inner_treespec, tree, /, is_leaf=None)[source]

Transform a tree having tree structure (outer, inner) into one having structure (inner, outer).

>>> outer_treespec = tree_structure({'a': 1, 'b': 2, 'c': (3, 4)})
>>> outer_treespec
PyTreeSpec({'a': *, 'b': *, 'c': (*, *)})
>>> inner_treespec = tree_structure((1, 2))
>>> inner_treespec
PyTreeSpec((*, *))
>>> tree = {'a': (1, 2), 'b': (3, 4), 'c': ((5, 6), (7, 8))}
>>> tree_transpose(outer_treespec, inner_treespec, tree)
({'a': 1, 'b': 3, 'c': (5, 7)}, {'a': 2, 'b': 4, 'c': (6, 8)})

For performance reasons, this function is only checks for the number of leaves in the input pytree, not the structure. The result is only enumerated up to the original order of leaves in tree, then transpose depends on the number of leaves in structure (inner, outer). The caller is responsible for ensuring that the input pytree has a prefix structure of outer_treespec followed by a prefix structure of inner_treespec. Otherwise, the result may be incorrect.

>>> tree_transpose(outer_treespec, inner_treespec, list(range(1, 9)))
({'a': 1, 'b': 3, 'c': (5, 7)}, {'a': 2, 'b': 4, 'c': (6, 8)})

Parameters:

outer_treespec (PyTreeSpec) – A treespec object representing the outer structure of the pytree.
inner_treespec (PyTreeSpec) – A treespec object representing the inner structure of the pytree.
tree (pytree) – A pytree to be transposed.
is_leaf (callable, optional) – An optionally specified function that will be called at each flattening step. It should return a boolean, with True stopping the traversal and the whole subtree being treated as a leaf, and False indicating the flattening should traverse the current object.

Return type:

PyTree [TypeVar(T)]

Returns:

A new pytree with the same structure as inner_treespec but with the value at each leaf has the same structure as outer_treespec.

optree.tree_transpose_map(func, tree, /, *rests, inner_treespec=None, is_leaf=None, none_is_leaf=False, namespace='')[source]

Map a multi-input function over pytree args to produce a new pytree with transposed structure.

>>> tree = {'b': (2, [3, 4]), 'a': 1, 'c': (5, 6)}
>>> tree_transpose_map(
...     lambda x: {'identity': x, 'double': 2 * x},
...     tree,
... )
{
    'identity': {'b': (2, [3, 4]), 'a': 1, 'c': (5, 6)},
    'double': {'b': (4, [6, 8]), 'a': 2, 'c': (10, 12)}
}
>>> tree_transpose_map(
...     lambda x: {'identity': x, 'double': (x, x)},
...     tree,
... )
{
    'identity': {'b': (2, [3, 4]), 'a': 1, 'c': (5, 6)},
    'double': (
        {'b': (2, [3, 4]), 'a': 1, 'c': (5, 6)},
        {'b': (2, [3, 4]), 'a': 1, 'c': (5, 6)}
    )
}
>>> tree_transpose_map(
...     lambda x: {'identity': x, 'double': (x, x)},
...     tree,
...     inner_treespec=tree_structure({'identity': 0, 'double': 0}),
... )
{
    'identity': {'b': (2, [3, 4]), 'a': 1, 'c': (5, 6)},
    'double': {'b': ((2, 2), [(3, 3), (4, 4)]), 'a': (1, 1), 'c': ((5, 5), (6, 6))}
}

Parameters:

func (callable) – A function that takes 1 + len(rests) arguments, to be applied at the corresponding leaves of the pytrees.
tree (pytree) – A pytree to be mapped over, with each leaf providing the first positional argument to function func.
rests (tuple of pytree) – A tuple of pytrees, each of which has the same structure as tree or has tree as a prefix.
inner_treespec (PyTreeSpec, optional) – The treespec object representing the inner structure of the result pytree. If not specified, the inner structure is inferred from the result of the function func on the first leaf. (default: None)
is_leaf (callable, optional) – An optionally specified function that will be called at each flattening step. It should return a boolean, with True stopping the traversal and the whole subtree being treated as a leaf, and False indicating the flattening should traverse the current object.
none_is_leaf (bool, optional) – Whether to treat None as a leaf. If False, None is a non-leaf node with arity 0. Thus None is contained in the treespec rather than in the leaves list and None will be remain in the result pytree. (default: False)
namespace (str, optional) – The registry namespace used for custom pytree node types. (default: '', i.e., the global namespace)

Return type:

PyTree [TypeVar(U)]

Returns:

A new nested pytree with the same structure as inner_treespec but with the value at each leaf has the same structure as tree. The subtree at each leaf is given by the result of function func(x, *xs) where x is the value at the corresponding leaf in tree and xs is the tuple of values at corresponding nodes in rests.

optree.tree_transpose_map_with_path(func, tree, /, *rests, inner_treespec=None, is_leaf=None, none_is_leaf=False, namespace='')[source]

Map a multi-input function over pytree args as well as the tree paths to produce a new pytree with transposed structure.

>>> tree = {'b': (2, [3, 4]), 'a': 1, 'c': (5, 6)}
>>> tree_transpose_map_with_path(
...     lambda p, x: {'depth': len(p), 'value': x},
...     tree,
... )
{
    'depth': {'b': (2, [3, 3]), 'a': 1, 'c': (2, 2)},
    'value': {'b': (2, [3, 4]), 'a': 1, 'c': (5, 6)}
}
>>> tree_transpose_map_with_path(
...     lambda p, x: {'path': p, 'value': x},
...     tree,
...     inner_treespec=tree_structure({'path': 0, 'value': 0}),
... )
{
    'path': {
        'b': (('b', 0), [('b', 1, 0), ('b', 1, 1)]),
        'a': ('a',),
        'c': (('c', 0), ('c', 1))
    },
    'value': {'b': (2, [3, 4]), 'a': 1, 'c': (5, 6)}
}

Parameters:

func (callable) – A function that takes 2 + len(rests) arguments, to be applied at the corresponding leaves of the pytrees with extra paths.
tree (pytree) – A pytree to be mapped over, with each leaf providing the second positional argument and the corresponding path providing the first positional argument to function func.
rests (tuple of pytree) – A tuple of pytrees, each of which has the same structure as tree or has tree as a prefix.
inner_treespec (PyTreeSpec, optional) – The treespec object representing the inner structure of the result pytree. If not specified, the inner structure is inferred from the result of the function func on the first leaf. (default: None)
is_leaf (callable, optional) – An optionally specified function that will be called at each flattening step. It should return a boolean, with True stopping the traversal and the whole subtree being treated as a leaf, and False indicating the flattening should traverse the current object.
none_is_leaf (bool, optional) – Whether to treat None as a leaf. If False, None is a non-leaf node with arity 0. Thus None is contained in the treespec rather than in the leaves list and None will be remain in the result pytree. (default: False)
namespace (str, optional) – The registry namespace used for custom pytree node types. (default: '', i.e., the global namespace)

Return type:

PyTree [TypeVar(U)]

Returns:

A new nested pytree with the same structure as inner_treespec but with the value at each leaf has the same structure as tree. The subtree at each leaf is given by the result of function func(p, x, *xs) where (p, x) are the path and value at the corresponding leaf in tree and xs is the tuple of values at corresponding nodes in rests.

optree.tree_transpose_map_with_accessor(func, tree, /, *rests, inner_treespec=None, is_leaf=None, none_is_leaf=False, namespace='')[source]

Map a multi-input function over pytree args as well as the tree accessors to produce a new pytree with transposed structure.

>>> tree = {'b': (2, [3, 4]), 'a': 1, 'c': (5, 6)}
>>> tree_transpose_map_with_accessor(
...     lambda a, x: {'depth': len(a), 'code': a.codify('tree'), 'value': x},
...     tree,
... )
{
    'depth': {
        'b': (2, [3, 3]),
        'a': 1,
        'c': (2, 2)
    },
    'code': {
        'b': ("tree['b'][0]", ["tree['b'][1][0]", "tree['b'][1][1]"]),
        'a': "tree['a']",
        'c': ("tree['c'][0]", "tree['c'][1]")
    },
    'value': {
        'b': (2, [3, 4]),
        'a': 1,
        'c': (5, 6)
    }
}
>>> tree_transpose_map_with_accessor(
...     lambda a, x: {'path': a.path, 'accessor': a, 'value': x},
...     tree,
...     inner_treespec=tree_structure({'path': 0, 'accessor': 0, 'value': 0}),
... )
{
    'path': {
        'b': (('b', 0), [('b', 1, 0), ('b', 1, 1)]),
        'a': ('a',),
        'c': (('c', 0), ('c', 1))
    },
    'accessor': {
        'b': (
            PyTreeAccessor(*['b'][0], ...),
            [
                PyTreeAccessor(*['b'][1][0], ...),
                PyTreeAccessor(*['b'][1][1], ...)
            ]
        ),
        'a': PyTreeAccessor(*['a'], ...),
        'c': (
            PyTreeAccessor(*['c'][0], ...),
            PyTreeAccessor(*['c'][1], ...)
        )
    },
    'value': {'b': (2, [3, 4]), 'a': 1, 'c': (5, 6)}
}

Parameters:

func (callable) – A function that takes 2 + len(rests) arguments, to be applied at the corresponding leaves of the pytrees with extra accessors.
tree (pytree) – A pytree to be mapped over, with each leaf providing the second positional argument and the corresponding path providing the first positional argument to function func.
rests (tuple of pytree) – A tuple of pytrees, each of which has the same structure as tree or has tree as a prefix.
inner_treespec (PyTreeSpec, optional) – The treespec object representing the inner structure of the result pytree. If not specified, the inner structure is inferred from the result of the function func on the first leaf. (default: None)
is_leaf (callable, optional) – An optionally specified function that will be called at each flattening step. It should return a boolean, with True stopping the traversal and the whole subtree being treated as a leaf, and False indicating the flattening should traverse the current object.
none_is_leaf (bool, optional) – Whether to treat None as a leaf. If False, None is a non-leaf node with arity 0. Thus None is contained in the treespec rather than in the leaves list and None will be remain in the result pytree. (default: False)
namespace (str, optional) – The registry namespace used for custom pytree node types. (default: '', i.e., the global namespace)

Return type:

PyTree [TypeVar(U)]

Returns:

A new nested pytree with the same structure as inner_treespec but with the value at each leaf has the same structure as tree. The subtree at each leaf is given by the result of function func(a, x, *xs) where (a, x) are the accessor and value at the corresponding leaf in tree and xs is the tuple of values at corresponding nodes in rests.

optree.tree_broadcast_prefix(prefix_tree, full_tree, /, is_leaf=None, *, none_is_leaf=False, namespace='')[source]

Return a pytree of same structure of full_tree with broadcasted subtrees in prefix_tree.

If a prefix_tree is a prefix of a full_tree, this means the full_tree can be constructed by replacing the leaves of prefix_tree with appropriate subtrees.

This function returns a pytree with the same size as full_tree. The leaves are replicated from prefix_tree. The number of replicas is determined by the corresponding subtree in full_tree.

>>> tree_broadcast_prefix(1, [2, 3, 4])
[1, 1, 1]
>>> tree_broadcast_prefix([1, 2, 3], [4, 5, 6])
[1, 2, 3]
>>> tree_broadcast_prefix([1, 2, 3], [4, 5, 6, 7])
Traceback (most recent call last):
    ...
ValueError: list arity mismatch; expected: 3, got: 4; list: [4, 5, 6, 7].
>>> tree_broadcast_prefix([1, 2, 3], [4, 5, (6, 7)])
[1, 2, (3, 3)]
>>> tree_broadcast_prefix([1, 2, 3], [4, 5, {'a': 6, 'b': 7, 'c': (None, 8)}])
[1, 2, {'a': 3, 'b': 3, 'c': (None, 3)}]
>>> tree_broadcast_prefix([1, 2, 3], [4, 5, {'a': 6, 'b': 7, 'c': (None, 8)}], none_is_leaf=True)
[1, 2, {'a': 3, 'b': 3, 'c': (3, 3)}]

Parameters:

prefix_tree (pytree) – A pytree with the prefix structure of full_tree.
full_tree (pytree) – A pytree with the suffix structure of prefix_tree.
is_leaf (callable, optional) – An optionally specified function that will be called at each flattening step. It should return a boolean, with True stopping the traversal and the whole subtree being treated as a leaf, and False indicating the flattening should traverse the current object.
none_is_leaf (bool, optional) – Whether to treat None as a leaf. If False, None is a non-leaf node with arity 0. Thus None is contained in the treespec rather than in the leaves list and None will be remain in the result pytree. (default: False)
namespace (str, optional) – The registry namespace used for custom pytree node types. (default: '', i.e., the global namespace)

Return type:

PyTree [TypeVar(T)]

Returns:

A pytree of same structure of full_tree with broadcasted subtrees in prefix_tree.

optree.broadcast_prefix(prefix_tree, full_tree, /, is_leaf=None, *, none_is_leaf=False, namespace='')[source]

Return a list of broadcasted leaves in prefix_tree to match the number of leaves in full_tree.

If a prefix_tree is a prefix of a full_tree, this means the full_tree can be constructed by replacing the leaves of prefix_tree with appropriate subtrees.

This function returns a list of leaves with the same size as full_tree. The leaves are replicated from prefix_tree. The number of replicas is determined by the corresponding subtree in full_tree.

>>> broadcast_prefix(1, [2, 3, 4])
[1, 1, 1]
>>> broadcast_prefix([1, 2, 3], [4, 5, 6])
[1, 2, 3]
>>> broadcast_prefix([1, 2, 3], [4, 5, 6, 7])
Traceback (most recent call last):
    ...
ValueError: list arity mismatch; expected: 3, got: 4; list: [4, 5, 6, 7].
>>> broadcast_prefix([1, 2, 3], [4, 5, (6, 7)])
[1, 2, 3, 3]
>>> broadcast_prefix([1, 2, 3], [4, 5, {'a': 6, 'b': 7, 'c': (None, 8)}])
[1, 2, 3, 3, 3]
>>> broadcast_prefix([1, 2, 3], [4, 5, {'a': 6, 'b': 7, 'c': (None, 8)}], none_is_leaf=True)
[1, 2, 3, 3, 3, 3]

Parameters:

prefix_tree (pytree) – A pytree with the prefix structure of full_tree.
full_tree (pytree) – A pytree with the suffix structure of prefix_tree.
is_leaf (callable, optional) – An optionally specified function that will be called at each flattening step. It should return a boolean, with True stopping the traversal and the whole subtree being treated as a leaf, and False indicating the flattening should traverse the current object.
none_is_leaf (bool, optional) – Whether to treat None as a leaf. If False, None is a non-leaf node with arity 0. Thus None is contained in the treespec rather than in the leaves list and None will be remain in the result pytree. (default: False)
namespace (str, optional) – The registry namespace used for custom pytree node types. (default: '', i.e., the global namespace)

Return type:

list[TypeVar(T)]

Returns:

A list of leaves in prefix_tree broadcasted to match the number of leaves in full_tree.

optree.tree_broadcast_common(tree, other_tree, /, is_leaf=None, *, none_is_leaf=False, namespace='')[source]

Return two pytrees of common suffix structure of tree and other_tree with broadcasted subtrees.

If a suffix_tree is a suffix of a tree, this means the suffix_tree can be constructed by replacing the leaves of tree with appropriate subtrees.

This function returns two pytrees with the same structure. The tree structure is the common suffix structure of tree and other_tree. The leaves are replicated from tree and other_tree. The number of replicas is determined by the corresponding subtree in the suffix structure.

>>> tree_broadcast_common(1, [2, 3, 4])
([1, 1, 1], [2, 3, 4])
>>> tree_broadcast_common([1, 2, 3], [4, 5, 6])
([1, 2, 3], [4, 5, 6])
>>> tree_broadcast_common([1, 2, 3], [4, 5, 6, 7])
Traceback (most recent call last):
    ...
ValueError: list arity mismatch; expected: 3, got: 4.
>>> tree_broadcast_common([1, (2, 3), 4], [5, 6, (7, 8)])
([1, (2, 3), (4, 4)], [5, (6, 6), (7, 8)])
>>> tree_broadcast_common([1, {'a': (2, 3)}, 4], [5, 6, {'a': 7, 'b': 8, 'c': (None, 9)}])
([1, {'a': (2, 3)}, {'a': 4, 'b': 4, 'c': (None, 4)}],
 [5, {'a': (6, 6)}, {'a': 7, 'b': 8, 'c': (None, 9)}])
>>> tree_broadcast_common([1, {'a': (2, 3)}, 4], [5, 6, {'a': 7, 'b': 8, 'c': (None, 9)}], none_is_leaf=True)
([1, {'a': (2, 3)}, {'a': 4, 'b': 4, 'c': (4, 4)}],
 [5, {'a': (6, 6)}, {'a': 7, 'b': 8, 'c': (None, 9)}])
>>> tree_broadcast_common([1, None], [None, 2])
([None, None], [None, None])
>>> tree_broadcast_common([1, None], [None, 2], none_is_leaf=True)
([1, None], [None, 2])

Parameters:

tree (pytree) – A pytree has a common suffix structure of other_tree.
other_tree (pytree) – A pytree has a common suffix structure of tree.
is_leaf (callable, optional) – An optionally specified function that will be called at each flattening step. It should return a boolean, with True stopping the traversal and the whole subtree being treated as a leaf, and False indicating the flattening should traverse the current object.
none_is_leaf (bool, optional) – Whether to treat None as a leaf. If False, None is a non-leaf node with arity 0. Thus None is contained in the treespec rather than in the leaves list and None will be remain in the result pytree. (default: False)
namespace (str, optional) – The registry namespace used for custom pytree node types. (default: '', i.e., the global namespace)

Return type:

Returns:

Two pytrees of common suffix structure of tree and other_tree with broadcasted subtrees.

optree.broadcast_common(tree, other_tree, /, is_leaf=None, *, none_is_leaf=False, namespace='')[source]

Return two lists of leaves in tree and other_tree broadcasted to match the number of leaves in the common suffix structure.

If a suffix_tree is a suffix of a tree, this means the suffix_tree can be constructed by replacing the leaves of tree with appropriate subtrees.

This function returns two pytrees with the same structure. The tree structure is the common suffix structure of tree and other_tree. The leaves are replicated from tree and other_tree. The number of replicas is determined by the corresponding subtree in the suffix structure.

>>> broadcast_common(1, [2, 3, 4])
([1, 1, 1], [2, 3, 4])
>>> broadcast_common([1, 2, 3], [4, 5, 6])
([1, 2, 3], [4, 5, 6])
>>> broadcast_common([1, 2, 3], [4, 5, 6, 7])
Traceback (most recent call last):
    ...
ValueError: list arity mismatch; expected: 3, got: 4.
>>> broadcast_common([1, (2, 3), 4], [5, 6, (7, 8)])
([1, 2, 3, 4, 4], [5, 6, 6, 7, 8])
>>> broadcast_common([1, {'a': (2, 3)}, 4], [5, 6, {'a': 7, 'b': 8, 'c': (None, 9)}])
([1, 2, 3, 4, 4, 4], [5, 6, 6, 7, 8, 9])
>>> broadcast_common([1, {'a': (2, 3)}, 4], [5, 6, {'a': 7, 'b': 8, 'c': (None, 9)}], none_is_leaf=True)
([1, 2, 3, 4, 4, 4, 4], [5, 6, 6, 7, 8, None, 9])
>>> broadcast_common([1, None], [None, 2])
([], [])
>>> broadcast_common([1, None], [None, 2], none_is_leaf=True)
([1, None], [None, 2])

Parameters:

tree (pytree) – A pytree has a common suffix structure of other_tree.
other_tree (pytree) – A pytree has a common suffix structure of tree.
is_leaf (callable, optional) – An optionally specified function that will be called at each flattening step. It should return a boolean, with True stopping the traversal and the whole subtree being treated as a leaf, and False indicating the flattening should traverse the current object.
none_is_leaf (bool, optional) – Whether to treat None as a leaf. If False, None is a non-leaf node with arity 0. Thus None is contained in the treespec rather than in the leaves list and None will be remain in the result pytree. (default: False)
namespace (str, optional) – The registry namespace used for custom pytree node types. (default: '', i.e., the global namespace)

Return type:

tuple[list[TypeVar(T)], list[TypeVar(T)]]

Returns:

Two lists of leaves in tree and other_tree broadcasted to match the number of leaves in the common suffix structure.

optree.tree_broadcast_map(func, tree, /, *rests, is_leaf=None, none_is_leaf=False, namespace='')[source]

Map a multi-input function over pytree args to produce a new pytree.

If only one input is provided, this function is the same as tree_map():

>>> tree_broadcast_map(lambda x: x + 1, {'x': 7, 'y': (42, 64)})
{'x': 8, 'y': (43, 65)}
>>> tree_broadcast_map(lambda x: x + 1, {'x': 7, 'y': (42, 64), 'z': None})
{'x': 8, 'y': (43, 65), 'z': None}
>>> tree_broadcast_map(lambda x: x is None, {'x': 7, 'y': (42, 64), 'z': None})
{'x': False, 'y': (False, False), 'z': None}
>>> tree_broadcast_map(lambda x: x is None, {'x': 7, 'y': (42, 64), 'z': None}, none_is_leaf=True)
{'x': False, 'y': (False, False), 'z': True}

If multiple inputs are given, all input trees will be broadcasted to the common suffix structure of all inputs:

>>> tree_broadcast_map(lambda x, y: x * y, [5, 6, (3, 4)], [{'a': 7, 'b': 9}, [1, 2], 8])
[{'a': 35, 'b': 45}, [6, 12], (24, 32)]

Parameters:

func (callable) – A function that takes 1 + len(rests) arguments, to be applied at the corresponding leaves of the pytrees.
tree (pytree) – A pytree to be mapped over, with each leaf providing the first positional argument to function func.
rests (tuple of pytree) – A tuple of pytrees, they should have a common suffix structure with each other and with tree.
is_leaf (callable, optional) – An optionally specified function that will be called at each flattening step. It should return a boolean, with True stopping the traversal and the whole subtree being treated as a leaf, and False indicating the flattening should traverse the current object.
none_is_leaf (bool, optional) – Whether to treat None as a leaf. If False, None is a non-leaf node with arity 0. Thus None is contained in the treespec rather than in the leaves list and None will be remain in the result pytree. (default: False)
namespace (str, optional) – The registry namespace used for custom pytree node types. (default: '', i.e., the global namespace)

Return type:

PyTree [TypeVar(U)]

Returns:

A new pytree with the structure as the common suffix structure of tree and rests but with the value at each leaf given by func(x, *xs) where x is the value at the corresponding leaf (may be broadcasted) in tree and xs is the tuple of values at corresponding leaves (may be broadcasted) in rests.

optree.tree_broadcast_map_with_path(func, tree, /, *rests, is_leaf=None, none_is_leaf=False, namespace='')[source]

Map a multi-input function over pytree args as well as the tree paths to produce a new pytree.

If only one input is provided, this function is the same as tree_map():

>>> tree_broadcast_map_with_path(lambda p, x: (len(p), x), {'x': 7, 'y': (42, 64)})
{'x': (1, 7), 'y': ((2, 42), (2, 64))}
>>> tree_broadcast_map_with_path(lambda p, x: x + len(p), {'x': 7, 'y': (42, 64), 'z': None})
{'x': 8, 'y': (44, 66), 'z': None}
>>> tree_broadcast_map_with_path(lambda p, x: p, {'x': 7, 'y': (42, 64), 'z': {1.5: None}})
{'x': ('x',), 'y': (('y', 0), ('y', 1)), 'z': {1.5: None}}
>>> tree_broadcast_map_with_path(lambda p, x: p, {'x': 7, 'y': (42, 64), 'z': {1.5: None}}, none_is_leaf=True)
{'x': ('x',), 'y': (('y', 0), ('y', 1)), 'z': {1.5: ('z', 1.5)}}

If multiple inputs are given, all input trees will be broadcasted to the common suffix structure of all inputs:

>>> tree_broadcast_map_with_path(
...     lambda p, x, y: (p, x * y),
...     [5, 6, (3, 4)],
...     [{'a': 7, 'b': 9}, [1, 2], 8],
... )
[
    {'a': ((0, 'a'), 35), 'b': ((0, 'b'), 45)},
    [((1, 0), 6), ((1, 1), 12)],
    (((2, 0), 24), ((2, 1), 32))
]

Parameters:

func (callable) – A function that takes 2 + len(rests) arguments, to be applied at the corresponding leaves of the pytrees with extra paths.
tree (pytree) – A pytree to be mapped over, with each leaf providing the first positional argument to function func.
rests (tuple of pytree) – A tuple of pytrees, they should have a common suffix structure with each other and with tree.
is_leaf (callable, optional) – An optionally specified function that will be called at each flattening step. It should return a boolean, with True stopping the traversal and the whole subtree being treated as a leaf, and False indicating the flattening should traverse the current object.
none_is_leaf (bool, optional) – Whether to treat None as a leaf. If False, None is a non-leaf node with arity 0. Thus None is contained in the treespec rather than in the leaves list and None will be remain in the result pytree. (default: False)
namespace (str, optional) – The registry namespace used for custom pytree node types. (default: '', i.e., the global namespace)

Return type:

PyTree [TypeVar(U)]

Returns:

A new pytree with the structure as the common suffix structure of tree and rests but with the value at each leaf given by func(p, x, *xs) where (p, x) are the path and value at the corresponding leaf (may be broadcasted) in and xs is the tuple of values at corresponding leaves (may be broadcasted) in rests.

optree.tree_broadcast_map_with_accessor(func, tree, /, *rests, is_leaf=None, none_is_leaf=False, namespace='')[source]

Map a multi-input function over pytree args as well as the tree accessors to produce a new pytree.

If only one input is provided, this function is the same as tree_map():

>>> tree_broadcast_map_with_accessor(lambda a, x: (len(a), x), {'x': 7, 'y': (42, 64)})
{'x': (1, 7), 'y': ((2, 42), (2, 64))}
>>> tree_broadcast_map_with_accessor(lambda a, x: x + len(a), {'x': 7, 'y': (42, 64), 'z': None})
{'x': 8, 'y': (44, 66), 'z': None}
>>> tree_broadcast_map_with_accessor(
...     lambda a, x: a.codify('tree'),
...     {'x': 7, 'y': (42, 64), 'z': {1.5: None}},
... )
{
    'x': "tree['x']",
    'y': ("tree['y'][0]", "tree['y'][1]"),
    'z': {1.5: None}
}
>>> tree_broadcast_map_with_accessor(
...     lambda a, x: a.codify('tree'),
...     {'x': 7, 'y': (42, 64), 'z': {1.5: None}},
...     none_is_leaf=True,
... )
{
    'x': "tree['x']",
    'y': ("tree['y'][0]", "tree['y'][1]"),
    'z': {1.5: "tree['z'][1.5]"}
}

If multiple inputs are given, all input trees will be broadcasted to the common suffix structure of all inputs:

>>> tree_broadcast_map_with_accessor(
...     lambda a, x, y: f'{a.codify("tree")} = {x * y}',
...     [5, 6, (3, 4)],
...     [{'a': 7, 'b': 9}, [1, 2], 8],
... )
[
    {'a': "tree[0]['a'] = 35", 'b': "tree[0]['b'] = 45"},
    ['tree[1][0] = 6', 'tree[1][1] = 12'],
    ('tree[2][0] = 24', 'tree[2][1] = 32')
]

Parameters:

func (callable) – A function that takes 2 + len(rests) arguments, to be applied at the corresponding leaves of the pytrees with extra accessors.
tree (pytree) – A pytree to be mapped over, with each leaf providing the first positional argument to function func.
rests (tuple of pytree) – A tuple of pytrees, they should have a common suffix structure with each other and with tree.
is_leaf (callable, optional) – An optionally specified function that will be called at each flattening step. It should return a boolean, with True stopping the traversal and the whole subtree being treated as a leaf, and False indicating the flattening should traverse the current object.
none_is_leaf (bool, optional) – Whether to treat None as a leaf. If False, None is a non-leaf node with arity 0. Thus None is contained in the treespec rather than in the leaves list and None will be remain in the result pytree. (default: False)
namespace (str, optional) – The registry namespace used for custom pytree node types. (default: '', i.e., the global namespace)

Return type:

PyTree [TypeVar(U)]

Returns:

A new pytree with the structure as the common suffix structure of tree and rests but with the value at each leaf given by func(a, x, *xs) where (a, x) are the accessor and value at the corresponding leaf (may be broadcasted) in and xs is the tuple of values at corresponding leaves (may be broadcasted) in rests.

optree.tree_reduce(func, tree, /, initial=<MISSING>, *, is_leaf=None, none_is_leaf=False, namespace='')[source]

Traversal through a pytree and reduce the leaves in left-to-right depth-first order.

See also tree_leaves() and tree_sum().

>>> tree_reduce(lambda x, y: x + y, {'x': 1, 'y': (2, 3)})
6
>>> tree_reduce(lambda x, y: x + y, {'x': 1, 'y': (2, None), 'z': 3})  # `None` is a non-leaf node with arity 0 by default
6
>>> tree_reduce(lambda x, y: x and y, {'x': 1, 'y': (2, None), 'z': 3})
3
>>> tree_reduce(lambda x, y: x and y, {'x': 1, 'y': (2, None), 'z': 3}, none_is_leaf=True)
None

Parameters:

func (callable) – A function that takes two arguments and returns a value of the same type.
tree (pytree) – A pytree to be traversed.
initial (object, optional) – An initial value to be used for the reduction. If not provided, the first leaf value is used as the initial value.
is_leaf (callable, optional) – An optionally specified function that will be called at each flattening step. It should return a boolean, with True stopping the traversal and the whole subtree being treated as a leaf, and False indicating the flattening should traverse the current object.
none_is_leaf (bool, optional) – Whether to treat None as a leaf. If False, None is a non-leaf node with arity 0. Thus None is contained in the treespec rather than in the leaves list and None will be remain in the result pytree. (default: False)
namespace (str, optional) – The registry namespace used for custom pytree node types. (default: '', i.e., the global namespace)

Return type:

TypeVar(T)

Returns:

The result of reducing the leaves of the pytree using func.

optree.tree_sum(tree, /, start=0, *, is_leaf=None, none_is_leaf=False, namespace='')[source]

Sum start and leaf values in tree in left-to-right depth-first order and return the total.

See also tree_leaves() and tree_reduce().

>>> tree_sum({'x': 1, 'y': (2, 3)})
6
>>> tree_sum({'x': 1, 'y': (2, None), 'z': 3})  # `None` is a non-leaf node with arity 0 by default
6
>>> tree_sum({'x': 1, 'y': (2, None), 'z': 3}, none_is_leaf=True)
Traceback (most recent call last):
    ...
TypeError: unsupported operand type(s) for +: 'int' and 'NoneType'
>>> tree_sum({'x': 'a', 'y': ('b', None), 'z': 'c'}, start='')
'abc'
>>> tree_sum({'x': [1], 'y': ([2], [None]), 'z': [3]}, start=[], is_leaf=lambda x: isinstance(x, list))
[1, 2, None, 3]

Parameters:

tree (pytree) – A pytree to be traversed.
start (object, optional) – An initial value to be used for the sum. (default: 0)
is_leaf (callable, optional) – An optionally specified function that will be called at each flattening step. It should return a boolean, with True stopping the traversal and the whole subtree being treated as a leaf, and False indicating the flattening should traverse the current object.
none_is_leaf (bool, optional) – Whether to treat None as a leaf. If False, None is a non-leaf node with arity 0. Thus None is contained in the treespec rather than in the leaves list and None will be remain in the result pytree. (default: False)
namespace (str, optional) – The registry namespace used for custom pytree node types. (default: '', i.e., the global namespace)

Return type:

TypeVar(T)

Returns:

The total sum of start and leaf values in tree.

optree.tree_max(tree, /, *, default=<MISSING>, key=None, is_leaf=None, none_is_leaf=False, namespace='')[source]

Return the maximum leaf value in tree.

See also tree_leaves() and tree_min().

>>> tree_max({})
Traceback (most recent call last):
    ...
ValueError: max() iterable argument is empty
>>> tree_max({}, default=0)
0
>>> tree_max({'x': 0, 'y': (2, 1)})
2
>>> tree_max({'x': 0, 'y': (2, 1)}, key=lambda x: -x)
0
>>> tree_max({'a': None})  # `None` is a non-leaf node with arity 0 by default
Traceback (most recent call last):
    ...
ValueError: max() iterable argument is empty
>>> tree_max({'a': None}, default=0)  # `None` is a non-leaf node with arity 0 by default
0
>>> tree_max({'a': None}, none_is_leaf=True)
None
>>> tree_max(None)  # `None` is a non-leaf node with arity 0 by default
Traceback (most recent call last):
    ...
ValueError: max() iterable argument is empty
>>> tree_max(None, default=0)
0
>>> tree_max(None, none_is_leaf=True)
None

Parameters:

tree (pytree) – A pytree to be traversed.
default (object, optional) – The default value to return if tree is empty. If the tree is empty and default is not specified, raise a ValueError.
key (callable or None, optional) – An one argument ordering function like that used for list.sort().
is_leaf (callable, optional) – An optionally specified function that will be called at each flattening step. It should return a boolean, with True stopping the traversal and the whole subtree being treated as a leaf, and False indicating the flattening should traverse the current object.
none_is_leaf (bool, optional) – Whether to treat None as a leaf. If False, None is a non-leaf node with arity 0. Thus None is contained in the treespec rather than in the leaves list and None will be remain in the result pytree. (default: False)
namespace (str, optional) – The registry namespace used for custom pytree node types. (default: '', i.e., the global namespace)

Return type:

TypeVar(T)

Returns:

The maximum leaf value in tree.

optree.tree_min(tree, /, *, default=<MISSING>, key=None, is_leaf=None, none_is_leaf=False, namespace='')[source]

Return the minimum leaf value in tree.

See also tree_leaves() and tree_max().

>>> tree_min({})
Traceback (most recent call last):
    ...
ValueError: min() iterable argument is empty
>>> tree_min({}, default=0)
0
>>> tree_min({'x': 0, 'y': (2, 1)})
0
>>> tree_min({'x': 0, 'y': (2, 1)}, key=lambda x: -x)
2
>>> tree_min({'a': None})  # `None` is a non-leaf node with arity 0 by default
Traceback (most recent call last):
    ...
ValueError: min() iterable argument is empty
>>> tree_min({'a': None}, default=0)  # `None` is a non-leaf node with arity 0 by default
0
>>> tree_min({'a': None}, none_is_leaf=True)
None
>>> tree_min(None)  # `None` is a non-leaf node with arity 0 by default
Traceback (most recent call last):
    ...
ValueError: min() iterable argument is empty
>>> tree_min(None, default=0)
0
>>> tree_min(None, none_is_leaf=True)
None

Parameters:

tree (pytree) – A pytree to be traversed.
default (object, optional) – The default value to return if tree is empty. If the tree is empty and default is not specified, raise a ValueError.
key (callable or None, optional) – An one argument ordering function like that used for list.sort().
is_leaf (callable, optional) – An optionally specified function that will be called at each flattening step. It should return a boolean, with True stopping the traversal and the whole subtree being treated as a leaf, and False indicating the flattening should traverse the current object.
none_is_leaf (bool, optional) – Whether to treat None as a leaf. If False, None is a non-leaf node with arity 0. Thus None is contained in the treespec rather than in the leaves list and None will be remain in the result pytree. (default: False)
namespace (str, optional) – The registry namespace used for custom pytree node types. (default: '', i.e., the global namespace)

Return type:

TypeVar(T)

Returns:

The minimum leaf value in tree.

optree.tree_all(tree, /, *, is_leaf=None, none_is_leaf=False, namespace='')[source]

Test whether all leaves in tree are true (or if tree is empty).

See also tree_leaves() and tree_any().

>>> tree_all({})
True
>>> tree_all({'x': 1, 'y': (2, 3)})
True
>>> tree_all({'x': 1, 'y': (2, None), 'z': 3})  # `None` is a non-leaf node by default
True
>>> tree_all({'x': 1, 'y': (2, None), 'z': 3}, none_is_leaf=True)
False
>>> tree_all(None)  # `None` is a non-leaf node by default
True
>>> tree_all(None, none_is_leaf=True)
False

Parameters:

tree (pytree) – A pytree to be traversed.
is_leaf (callable, optional) – An optionally specified function that will be called at each flattening step. It should return a boolean, with True stopping the traversal and the whole subtree being treated as a leaf, and False indicating the flattening should traverse the current object.
none_is_leaf (bool, optional) – Whether to treat None as a leaf. If False, None is a non-leaf node with arity 0. Thus None is contained in the treespec rather than in the leaves list and None will be remain in the result pytree. (default: False)
namespace (str, optional) – The registry namespace used for custom pytree node types. (default: '', i.e., the global namespace)

Return type:

bool

Returns:

True if all leaves in tree are true, or if tree is empty. Otherwise, False.

optree.tree_any(tree, /, *, is_leaf=None, none_is_leaf=False, namespace='')[source]

Test whether all leaves in tree are true (or False if tree is empty).

See also tree_leaves() and tree_all().

>>> tree_any({})
False
>>> tree_any({'x': 0, 'y': (2, 0)})
True
>>> tree_any({'a': None})  # `None` is a non-leaf node with arity 0 by default
False
>>> tree_any({'a': None}, none_is_leaf=True)  # `None` is evaluated as false
False
>>> tree_any(None)  # `None` is a non-leaf node with arity 0 by default
False
>>> tree_any(None, none_is_leaf=True)  # `None` is evaluated as false
False

Parameters:

tree (pytree) – A pytree to be traversed.
is_leaf (callable, optional) – An optionally specified function that will be called at each flattening step. It should return a boolean, with True stopping the traversal and the whole subtree being treated as a leaf, and False indicating the flattening should traverse the current object.
none_is_leaf (bool, optional) – Whether to treat None as a leaf. If False, None is a non-leaf node with arity 0. Thus None is contained in the treespec rather than in the leaves list and None will be remain in the result pytree. (default: False)
namespace (str, optional) – The registry namespace used for custom pytree node types. (default: '', i.e., the global namespace)

Return type:

bool

Returns:

True if any leaves in tree are true, otherwise, False. If tree is empty, return False.

optree.tree_flatten_one_level(tree, /, is_leaf=None, *, none_is_leaf=False, namespace='')[source]

Flatten the pytree one level, returning a 4-tuple of children, metadata, path entries, and an unflatten function.

>>> children, metadata, entries, unflatten_func = tree_flatten_one_level({'b': (2, [3, 4]), 'a': 1, 'c': None, 'd': 5})
>>> children, metadata, entries
([1, (2, [3, 4]), None, 5], ['a', 'b', 'c', 'd'], ('a', 'b', 'c', 'd'))
>>> unflatten_func(metadata, children)
{'a': 1, 'b': (2, [3, 4]), 'c': None, 'd': 5}
>>> children, metadata, entries, unflatten_func = tree_flatten_one_level([{'a': 1, 'b': (2, 3)}, (4, 5)])
>>> children, metadata, entries
([{'a': 1, 'b': (2, 3)}, (4, 5)], None, (0, 1))
>>> unflatten_func(metadata, children)
[{'a': 1, 'b': (2, 3)}, (4, 5)]

Parameters:

tree (pytree) – A pytree to be traversed.
is_leaf (callable, optional) – An optionally specified function that will be called at each flattening step. It should return a boolean, with True stopping the traversal and the whole subtree being treated as a leaf, and False indicating the flattening should traverse the current object.
none_is_leaf (bool, optional) – Whether to treat None as a leaf. If False, None is a non-leaf node with arity 0. Thus None is contained in the treespec rather than in the leaves list and None will be remain in the result pytree. (default: False)
namespace (str, optional) – The registry namespace used for custom pytree node types. (default: '', i.e., the global namespace)

Return type:

FlattenOneLevelOutputEx[TypeVar(T)]

Returns:

A 4-tuple (children, metadata, entries, unflatten_func). The first element is a list of one-level children of the pytree node. The second element is the metadata used to reconstruct the pytree node. The third element is a tuple of path entries to the children. The fourth element is a function that can be used to unflatten the metadata and children back to the pytree node.

optree.prefix_errors(prefix_tree, full_tree, /, is_leaf=None, *, none_is_leaf=False, namespace='')[source]

Return a list of errors that would be raised by broadcast_prefix().

Return type:: list[Callable[[str], ValueError]]

optree.treespec_paths(treespec, /)[source]

Return a list of paths to the leaves of a treespec.

Return type:: list[tuple[Any, ...]]

>>> treespec = tree_structure({'b': 3, 'a': (0, [1, 2]), 'c': (4, None)})
>>> treespec
PyTreeSpec({'a': (*, [*, *]), 'b': *, 'c': (*, None)})
>>> treespec_paths(treespec)
[('a', 0), ('a', 1, 0), ('a', 1, 1), ('b',), ('c', 0)]

optree.treespec_accessors(treespec, /)[source]

Return a list of accessors to the leaves of a treespec.

Return type:: list[PyTreeAccessor]

>>> treespec = tree_structure({'b': 3, 'a': (0, [1, 2]), 'c': (4, None)})
>>> treespec
PyTreeSpec({'a': (*, [*, *]), 'b': *, 'c': (*, None)})
>>> treespec_accessors(treespec)
[
    PyTreeAccessor(*['a'][0], ...),
    PyTreeAccessor(*['a'][1][0], ...),
    PyTreeAccessor(*['a'][1][1], ...),
    PyTreeAccessor(*['b'], ...),
    PyTreeAccessor(*['c'][0], ...)
]
>>> treespec_accessors(treespec_leaf())
[PyTreeAccessor(*, ())]
>>> treespec_accessors(treespec_none())
[]

optree.treespec_entries(treespec, /)[source]

Return a list of one-level entries of a treespec to its children.

Return type:: list[Any]

>>> treespec = tree_structure({'b': 3, 'a': (0, [1, 2]), 'c': (4, None)})
>>> treespec
PyTreeSpec({'a': (*, [*, *]), 'b': *, 'c': (*, None)})
>>> treespec_entries(treespec)
['a', 'b', 'c']

optree.treespec_entry(treespec, index, /)[source]

Return the entry of a treespec at the given index.

Return type:: Any

optree.treespec_children(treespec, /)[source]

Return a list of treespecs for the children of a treespec.

Return type:: list[PyTreeSpec]

>>> treespec = tree_structure({'b': 3, 'a': (0, [1, 2]), 'c': (4, None)})
>>> treespec
PyTreeSpec({'a': (*, [*, *]), 'b': *, 'c': (*, None)})
>>> treespec_children(treespec)
[PyTreeSpec((*, [*, *])), PyTreeSpec(*), PyTreeSpec((*, None))]

optree.treespec_child(treespec, index, /)[source]

Return the treespec of the child of a treespec at the given index.

Return type:: PyTreeSpec

optree.treespec_one_level(treespec, /)[source]

Return the one-level tree structure of the treespec or None if the treespec is a leaf.

Return type:: PyTreeSpec | None

>>> treespec = tree_structure({'b': 3, 'a': (0, [1, 2]), 'c': (4, None)})
>>> treespec
PyTreeSpec({'a': (*, [*, *]), 'b': *, 'c': (*, None)})
>>> treespec_one_level(treespec)
PyTreeSpec({'a': *, 'b': *, 'c': *})

optree.treespec_transform(treespec, /, f_node=None, f_leaf=None)[source]

Transform a treespec by applying functions to its nodes and leaves.

Return type:: PyTreeSpec

>>> treespec = tree_structure({'b': 3, 'a': (0, [1, 2]), 'c': (4, None)})
>>> treespec
PyTreeSpec({'a': (*, [*, *]), 'b': *, 'c': (*, None)})
>>> treespec_transform(treespec, lambda spec: treespec_dict(zip(spec.entries(), spec.children())))
PyTreeSpec({'a': {0: *, 1: {0: *, 1: *}}, 'b': *, 'c': {0: *, 1: {}}})
>>> treespec_transform(
...     treespec,
...     lambda spec: (
...         treespec_ordereddict(zip(spec.entries(), spec.children()))
...         if spec.type is dict
...         else spec
...     ),
... )
PyTreeSpec(OrderedDict({'a': (*, [*, *]), 'b': *, 'c': (*, None)}))
>>> treespec_transform(
...     treespec,
...     lambda spec: (
...         treespec_ordereddict(tree_unflatten(spec, spec.children()))
...         if spec.type is dict
...         else spec
...     ),
... )
PyTreeSpec(OrderedDict({'b': (*, [*, *]), 'a': *, 'c': (*, None)}))
>>> treespec_transform(treespec, lambda spec: treespec_tuple(spec.children()))
PyTreeSpec(((*, (*, *)), *, (*, ())))
>>> treespec_transform(
...     treespec,
...     lambda spec: (
...         treespec_list(spec.children())
...         if spec.type is tuple
...         else spec
...     ),
... )
PyTreeSpec({'a': [*, [*, *]], 'b': *, 'c': [*, None]})
>>> treespec_transform(treespec, None, lambda spec: tree_structure((1, [2])))
PyTreeSpec({'a': ((*, [*]), [(*, [*]), (*, [*])]), 'b': (*, [*]), 'c': ((*, [*]), None)})

optree.treespec_is_leaf(treespec, /, *, strict=True)[source]

Return whether the treespec is a leaf that has no children.

This function is equivalent to treespec.is_leaf(strict=strict). If strict=False, it will return True if and only if the treespec represents a strict leaf. If strict=False, it will return True if the treespec represents a strict leaf or None or an empty container (e.g., an empty tuple).

>>> treespec_is_leaf(tree_structure(1))
True
>>> treespec_is_leaf(tree_structure((1, 2)))
False
>>> treespec_is_leaf(tree_structure(None))
False
>>> treespec_is_leaf(tree_structure(None), strict=False)
True
>>> treespec_is_leaf(tree_structure(None, none_is_leaf=False))
False
>>> treespec_is_leaf(tree_structure(None, none_is_leaf=True))
True
>>> treespec_is_leaf(tree_structure(()))
False
>>> treespec_is_leaf(tree_structure(()), strict=False)
True
>>> treespec_is_leaf(tree_structure([]))
False
>>> treespec_is_leaf(tree_structure([]), strict=False)
True

Parameters:

treespec (PyTreeSpec) – A treespec.
strict (bool, optional) – Whether not to treat None or an empty container (e.g., an empty tuple) as a leaf. (default: True)

Return type:

bool

Returns:

True if the treespec represents a leaf that has no children, otherwise, False.

optree.treespec_is_strict_leaf(treespec, /)[source]

Return whether the treespec is a strict leaf.

This function respects the none_is_leaf setting in the treespec. It is equivalent to treespec.is_leaf(strict=True). It will return True if and only if the treespec represents a strict leaf.

>>> treespec_is_strict_leaf(tree_structure(1))
True
>>> treespec_is_strict_leaf(tree_structure((1, 2)))
False
>>> treespec_is_strict_leaf(tree_structure(None))
False
>>> treespec_is_strict_leaf(tree_structure(None, none_is_leaf=False))
False
>>> treespec_is_strict_leaf(tree_structure(None, none_is_leaf=True))
True
>>> treespec_is_strict_leaf(tree_structure(()))
False
>>> treespec_is_strict_leaf(tree_structure([]))
False

Parameters:: treespec (PyTreeSpec) – A treespec.
Return type:: bool
Returns:: True if the treespec represents a strict leaf, otherwise, False.

optree.treespec_is_one_level(treespec, /)[source]

Return whether the treespec is a one-level tree structure.

Return type:: bool

>>> treespec_is_one_level(tree_structure(1))
False
>>> treespec_is_one_level(tree_structure((1, 2)))
True
>>> treespec_is_one_level(tree_structure({'a': 1, 'b': 2, 'c': 3}))
True
>>> treespec_is_one_level(tree_structure({'a': 1, 'b': (2, 3), 'c': 4}))
False
>>> treespec_is_one_level(tree_structure(None))
True

optree.treespec_is_prefix(treespec, other_treespec, /, *, strict=False)[source]

Return whether treespec is a prefix of other_treespec.

Return type:: bool

optree.treespec_is_suffix(treespec, other_treespec, /, *, strict=False)[source]

Return whether treespec is a suffix of other_treespec.

Return type:: bool

optree.treespec_leaf(*, none_is_leaf=False, namespace='')[source]

Make a treespec representing a leaf node.

>>> treespec_leaf()
PyTreeSpec(*)
>>> treespec_leaf(none_is_leaf=True)
PyTreeSpec(*, NoneIsLeaf)
>>> treespec_leaf(none_is_leaf=False) == treespec_leaf(none_is_leaf=True)
False
>>> treespec_leaf() == tree_structure(1)
True
>>> treespec_leaf(none_is_leaf=True) == tree_structure(1, none_is_leaf=True)
True
>>> treespec_leaf(none_is_leaf=True) == tree_structure(None, none_is_leaf=True)
True
>>> treespec_leaf(none_is_leaf=True) == tree_structure(None, none_is_leaf=False)
False
>>> treespec_leaf(none_is_leaf=True) == treespec_none(none_is_leaf=True)
True
>>> treespec_leaf(none_is_leaf=True) == treespec_none(none_is_leaf=False)
False
>>> treespec_leaf(none_is_leaf=False) == treespec_none(none_is_leaf=True)
False
>>> treespec_leaf(none_is_leaf=False) == treespec_none(none_is_leaf=False)
False

Parameters:

none_is_leaf (bool, optional) – Whether to treat None as a leaf. If False, None is a non-leaf node with arity 0. Thus None is contained in the treespec rather than in the leaves list and None will be remain in the result pytree. (default: False)
namespace (str, optional) – The registry namespace used for custom pytree node types. (default: '', i.e., the global namespace)

Return type:

PyTreeSpec

Returns:

A treespec representing a leaf node.

optree.treespec_none(*, none_is_leaf=False, namespace='')[source]

Make a treespec representing a None node.

>>> treespec_none()
PyTreeSpec(None)
>>> treespec_none(none_is_leaf=True)
PyTreeSpec(*, NoneIsLeaf)
>>> treespec_none(none_is_leaf=False) == treespec_none(none_is_leaf=True)
False
>>> treespec_none() == tree_structure(None)
True
>>> treespec_none() == tree_structure(1)
False
>>> treespec_none(none_is_leaf=True) == tree_structure(1, none_is_leaf=True)
True
>>> treespec_none(none_is_leaf=True) == tree_structure(None, none_is_leaf=True)
True
>>> treespec_none(none_is_leaf=True) == tree_structure(None, none_is_leaf=False)
False
>>> treespec_none(none_is_leaf=True) == treespec_leaf(none_is_leaf=True)
True
>>> treespec_none(none_is_leaf=False) == treespec_leaf(none_is_leaf=True)
False
>>> treespec_none(none_is_leaf=True) == treespec_leaf(none_is_leaf=False)
False
>>> treespec_none(none_is_leaf=False) == treespec_leaf(none_is_leaf=False)
False

Parameters:

none_is_leaf (bool, optional) – Whether to treat None as a leaf. If False, None is a non-leaf node with arity 0. Thus None is contained in the treespec rather than in the leaves list and None will be remain in the result pytree. (default: False)
namespace (str, optional) – The registry namespace used for custom pytree node types. (default: '', i.e., the global namespace)

Return type:

PyTreeSpec

Returns:

A treespec representing a None node.

optree.treespec_tuple(iterable=(), /, *, none_is_leaf=False, namespace='')[source]

Make a tuple treespec from an iterable of child treespecs.

>>> treespec_tuple([treespec_leaf(), treespec_leaf()])
PyTreeSpec((*, *))
>>> treespec_tuple([treespec_leaf(), treespec_leaf(), treespec_none()])
PyTreeSpec((*, *, None))
>>> treespec_tuple()
PyTreeSpec(())
>>> treespec_tuple([treespec_leaf(), treespec_tuple([treespec_leaf(), treespec_leaf()])])
PyTreeSpec((*, (*, *)))
>>> treespec_tuple([treespec_leaf(), tree_structure({'a': 1, 'b': 2})])
PyTreeSpec((*, {'a': *, 'b': *}))
>>> treespec_tuple([treespec_leaf(), tree_structure({'a': 1, 'b': 2}, none_is_leaf=True)])
Traceback (most recent call last):
    ...
ValueError: Expected treespec(s) with `none_is_leaf=False`.

Parameters:

iterable (iterable of PyTreeSpec, optional) – A iterable of child treespecs. They must have the same none_is_leaf and namespace values.
none_is_leaf (bool, optional) – Whether to treat None as a leaf. If False, None is a non-leaf node with arity 0. Thus None is contained in the treespec rather than in the leaves list and None will be remain in the result pytree. (default: False)
namespace (str, optional) – The registry namespace used for custom pytree node types. (default: '', i.e., the global namespace)

Return type:

PyTreeSpec

Returns:

A treespec representing a tuple node with the given children.

optree.treespec_list(iterable=(), /, *, none_is_leaf=False, namespace='')[source]

Make a list treespec from an iterable of child treespecs.

>>> treespec_list([treespec_leaf(), treespec_leaf()])
PyTreeSpec([*, *])
>>> treespec_list([treespec_leaf(), treespec_leaf(), treespec_none()])
PyTreeSpec([*, *, None])
>>> treespec_list()
PyTreeSpec([])
>>> treespec_list([treespec_leaf(), treespec_tuple([treespec_leaf(), treespec_leaf()])])
PyTreeSpec([*, (*, *)])
>>> treespec_list([treespec_leaf(), tree_structure({'a': 1, 'b': 2})])
PyTreeSpec([*, {'a': *, 'b': *}])
>>> treespec_list([treespec_leaf(), tree_structure({'a': 1, 'b': 2}, none_is_leaf=True)])
Traceback (most recent call last):
    ...
ValueError: Expected treespec(s) with `none_is_leaf=False`.

Parameters:

iterable (iterable of PyTreeSpec, optional) – A iterable of child treespecs. They must have the same none_is_leaf and namespace values.
none_is_leaf (bool, optional) – Whether to treat None as a leaf. If False, None is a non-leaf node with arity 0. Thus None is contained in the treespec rather than in the leaves list and None will be remain in the result pytree. (default: False)
namespace (str, optional) – The registry namespace used for custom pytree node types. (default: '', i.e., the global namespace)

Return type:

PyTreeSpec

Returns:

A treespec representing a list node with the given children.

optree.treespec_dict(mapping=(), /, *, none_is_leaf=False, namespace='', **kwargs)[source]

Make a dict treespec from a dict of child treespecs.

>>> treespec_dict({'a': treespec_leaf(), 'b': treespec_leaf()})
PyTreeSpec({'a': *, 'b': *})
>>> treespec_dict([('b', treespec_leaf()), ('c', treespec_leaf()), ('a', treespec_none())])
PyTreeSpec({'a': None, 'b': *, 'c': *})
>>> treespec_dict()
PyTreeSpec({})
>>> treespec_dict(a=treespec_leaf(), b=treespec_tuple([treespec_leaf(), treespec_leaf()]))
PyTreeSpec({'a': *, 'b': (*, *)})
>>> treespec_dict({'a': treespec_leaf(), 'b': tree_structure([1, 2])})
PyTreeSpec({'a': *, 'b': [*, *]})
>>> treespec_dict({'a': treespec_leaf(), 'b': tree_structure([1, 2], none_is_leaf=True)})
Traceback (most recent call last):
    ...
ValueError: Expected treespec(s) with `none_is_leaf=False`.

Parameters:

mapping (mapping of PyTreeSpec, optional) – A mapping of child treespecs. They must have the same none_is_leaf and namespace values.
none_is_leaf (bool, optional) – Whether to treat None as a leaf. If False, None is a non-leaf node with arity 0. Thus None is contained in the treespec rather than in the leaves list and None will be remain in the result pytree. (default: False)
namespace (str, optional) – The registry namespace used for custom pytree node types. (default: '', i.e., the global namespace)
**kwargs (PyTreeSpec, optional) – Additional child treespecs to add to the mapping.

Return type:

PyTreeSpec

Returns:

A treespec representing a dict node with the given children.

optree.treespec_namedtuple(namedtuple, /, *, none_is_leaf=False, namespace='')[source]

Make a namedtuple treespec from a namedtuple of child treespecs.

>>> from collections import namedtuple
>>> Point = namedtuple('Point', ['x', 'y'])
>>> treespec_namedtuple(Point(x=treespec_leaf(), y=treespec_leaf()))
PyTreeSpec(Point(x=*, y=*))
>>> treespec_namedtuple(Point(x=treespec_leaf(), y=treespec_tuple([treespec_leaf(), treespec_leaf()])))
PyTreeSpec(Point(x=*, y=(*, *)))
>>> treespec_namedtuple(Point(x=treespec_leaf(), y=tree_structure([1, 2])))
PyTreeSpec(Point(x=*, y=[*, *]))
>>> treespec_namedtuple(Point(x=treespec_leaf(), y=tree_structure([1, 2], none_is_leaf=True)))
Traceback (most recent call last):
    ...
ValueError: Expected treespec(s) with `none_is_leaf=False`.

Parameters:

namedtuple (namedtuple of PyTreeSpec) – A namedtuple of child treespecs. They must have the same none_is_leaf and namespace values.
none_is_leaf (bool, optional) – Whether to treat None as a leaf. If False, None is a non-leaf node with arity 0. Thus None is contained in the treespec rather than in the leaves list and None will be remain in the result pytree. (default: False)
namespace (str, optional) – The registry namespace used for custom pytree node types. (default: '', i.e., the global namespace)

Returns:

A treespec representing a dict node with the given children.

optree.treespec_ordereddict(mapping=(), /, *, none_is_leaf=False, namespace='', **kwargs)[source]

Make an OrderedDict treespec from an OrderedDict of child treespecs.

>>> treespec_ordereddict({'a': treespec_leaf(), 'b': treespec_leaf()})
PyTreeSpec(OrderedDict({'a': *, 'b': *}))
>>> treespec_ordereddict([('b', treespec_leaf()), ('c', treespec_leaf()), ('a', treespec_none())])
PyTreeSpec(OrderedDict({'b': *, 'c': *, 'a': None}))
>>> treespec_ordereddict()
PyTreeSpec(OrderedDict())
>>> treespec_ordereddict(a=treespec_leaf(), b=treespec_tuple([treespec_leaf(), treespec_leaf()]))
PyTreeSpec(OrderedDict({'a': *, 'b': (*, *)}))
>>> treespec_ordereddict({'a': treespec_leaf(), 'b': tree_structure([1, 2])})
PyTreeSpec(OrderedDict({'a': *, 'b': [*, *]}))
>>> treespec_ordereddict({'a': treespec_leaf(), 'b': tree_structure([1, 2], none_is_leaf=True)})
Traceback (most recent call last):
    ...
ValueError: Expected treespec(s) with `none_is_leaf=False`.

Parameters:

mapping (mapping of PyTreeSpec, optional) – A mapping of child treespecs. They must have the same none_is_leaf and namespace values.
none_is_leaf (bool, optional) – Whether to treat None as a leaf. If False, None is a non-leaf node with arity 0. Thus None is contained in the treespec rather than in the leaves list and None will be remain in the result pytree. (default: False)
namespace (str, optional) – The registry namespace used for custom pytree node types. (default: '', i.e., the global namespace)
**kwargs (PyTreeSpec, optional) – Additional child treespecs to add to the mapping.

Return type:

PyTreeSpec

Returns:

A treespec representing an OrderedDict node with the given children.

optree.treespec_defaultdict(default_factory=None, mapping=(), /, *, none_is_leaf=False, namespace='', **kwargs)[source]

Make a defaultdict treespec from a defaultdict of child treespecs.

>>> treespec_defaultdict(int, {'a': treespec_leaf(), 'b': treespec_leaf()})
PyTreeSpec(defaultdict(<class 'int'>, {'a': *, 'b': *}))
>>> treespec_defaultdict(int, [('b', treespec_leaf()), ('c', treespec_leaf()), ('a', treespec_none())])
PyTreeSpec(defaultdict(<class 'int'>, {'a': None, 'b': *, 'c': *}))
>>> treespec_defaultdict()
PyTreeSpec(defaultdict(None, {}))
>>> treespec_defaultdict(int)
PyTreeSpec(defaultdict(<class 'int'>, {}))
>>> treespec_defaultdict(int, a=treespec_leaf(), b=treespec_tuple([treespec_leaf(), treespec_leaf()]))
PyTreeSpec(defaultdict(<class 'int'>, {'a': *, 'b': (*, *)}))
>>> treespec_defaultdict(int, {'a': treespec_leaf(), 'b': tree_structure([1, 2])})
PyTreeSpec(defaultdict(<class 'int'>, {'a': *, 'b': [*, *]}))
>>> treespec_defaultdict(int, {'a': treespec_leaf(), 'b': tree_structure([1, 2], none_is_leaf=True)})
Traceback (most recent call last):
    ...
ValueError: Expected treespec(s) with `none_is_leaf=False`.

Parameters:

default_factory (callable or None, optional) – A factory function that will be used to create a missing value. (default: None)
mapping (mapping of PyTreeSpec, optional) – A mapping of child treespecs. They must have the same none_is_leaf and namespace values.
none_is_leaf (bool, optional) – Whether to treat None as a leaf. If False, None is a non-leaf node with arity 0. Thus None is contained in the treespec rather than in the leaves list and None will be remain in the result pytree. (default: False)
namespace (str, optional) – The registry namespace used for custom pytree node types. (default: '', i.e., the global namespace)
**kwargs (PyTreeSpec, optional) – Additional child treespecs to add to the mapping.

Return type:

PyTreeSpec

Returns:

A treespec representing a defaultdict node with the given children.

optree.treespec_deque(iterable=(), /, maxlen=None, *, none_is_leaf=False, namespace='')[source]

Make a deque treespec from a deque of child treespecs.

>>> treespec_deque([treespec_leaf(), treespec_leaf()])
PyTreeSpec(deque([*, *]))
>>> treespec_deque([treespec_leaf(), treespec_leaf(), treespec_none()], maxlen=5)
PyTreeSpec(deque([*, *, None], maxlen=5))
>>> treespec_deque()
PyTreeSpec(deque([]))
>>> treespec_deque([treespec_leaf(), treespec_tuple([treespec_leaf(), treespec_leaf()])])
PyTreeSpec(deque([*, (*, *)]))
>>> treespec_deque([treespec_leaf(), tree_structure({'a': 1, 'b': 2})], maxlen=5)
PyTreeSpec(deque([*, {'a': *, 'b': *}], maxlen=5))
>>> treespec_deque([treespec_leaf(), tree_structure({'a': 1, 'b': 2}, none_is_leaf=True)], maxlen=5)
Traceback (most recent call last):
    ...
ValueError: Expected treespec(s) with `none_is_leaf=False`.

Parameters:

iterable (iterable of PyTreeSpec, optional) – A iterable of child treespecs. They must have the same none_is_leaf and namespace values.
maxlen (int or None, optional) – The maximum size of a deque or None if unbounded. (default: None)
none_is_leaf (bool, optional) – Whether to treat None as a leaf. If False, None is a non-leaf node with arity 0. Thus None is contained in the treespec rather than in the leaves list and None will be remain in the result pytree. (default: False)
namespace (str, optional) – The registry namespace used for custom pytree node types. (default: '', i.e., the global namespace)

Return type:

PyTreeSpec

Returns:

A treespec representing a deque node with the given children.

optree.treespec_structseq(structseq, /, *, none_is_leaf=False, namespace='')[source]

Make a PyStructSequence treespec from a PyStructSequence of child treespecs.

Parameters:

structseq (PyStructSequence of PyTreeSpec) – A PyStructSequence of child treespecs. They must have the same none_is_leaf and namespace values.
none_is_leaf (bool, optional) – Whether to treat None as a leaf. If False, None is a non-leaf node with arity 0. Thus None is contained in the treespec rather than in the leaves list and None will be remain in the result pytree. (default: False)
namespace (str, optional) – The registry namespace used for custom pytree node types. (default: '', i.e., the global namespace)

Return type:

PyTreeSpec

Returns:

A treespec representing a PyStructSequence node with the given children.

optree.treespec_from_collection(collection, /, *, none_is_leaf=False, namespace='')[source]

Make a treespec from a collection of child treespecs.

>>> treespec_from_collection(None)
PyTreeSpec(None)
>>> treespec_from_collection(None, none_is_leaf=True)
PyTreeSpec(*, NoneIsLeaf)
>>> treespec_from_collection(object())
PyTreeSpec(*)
>>> treespec_from_collection([treespec_leaf(), treespec_none()])
PyTreeSpec([*, None])
>>> treespec_from_collection({'a': treespec_leaf(), 'b': treespec_none()})
PyTreeSpec({'a': *, 'b': None})
>>> treespec_from_collection(deque([treespec_leaf(), tree_structure({'a': 1, 'b': 2})], maxlen=5))
PyTreeSpec(deque([*, {'a': *, 'b': *}], maxlen=5))
>>> treespec_from_collection({'a': treespec_leaf(), 'b': (treespec_leaf(), treespec_none())})
Traceback (most recent call last):
    ...
ValueError: Expected a(n) dict of PyTreeSpec(s), got {'a': PyTreeSpec(*), 'b': (PyTreeSpec(*), PyTreeSpec(None))}.
>>> treespec_from_collection([treespec_leaf(), tree_structure({'a': 1, 'b': 2}, none_is_leaf=True)])
Traceback (most recent call last):
    ...
ValueError: Expected treespec(s) with `none_is_leaf=False`.

Parameters:

collection (collection of PyTreeSpec) – A collection of child treespecs. They must have the same none_is_leaf and namespace values.
none_is_leaf (bool, optional) – Whether to treat None as a leaf. If False, None is a non-leaf node with arity 0. Thus None is contained in the treespec rather than in the leaves list and None will be remain in the result pytree. (default: False)
namespace (str, optional) – The registry namespace used for custom pytree node types. (default: '', i.e., the global namespace)

Return type:

PyTreeSpec

Returns:

A treespec representing the same structure of the collection with the given children.

class optree.PyTreeEntry(entry, type, kind)

Bases: object

Base class for path entries.

entry: Any

type: builtins.type

kind: PyTreeKind

__add__(other, /)[source]

Join the path entry with another path entry or accessor.

Return type:: PyTreeAccessor

__annotations__ = {'entry': 'Any', 'kind': 'PyTreeKind', 'type': 'builtins.type'}

__call__(obj, /)[source]

Get the child object.

Return type:: Any

__dataclass_fields__ = {'entry': Field(name='entry',type='Any',default=<dataclasses._MISSING_TYPE object>,default_factory=<dataclasses._MISSING_TYPE object>,init=True,repr=True,hash=None,compare=True,metadata=mappingproxy({}),kw_only=False,_field_type=_FIELD), 'kind': Field(name='kind',type='PyTreeKind',default=<dataclasses._MISSING_TYPE object>,default_factory=<dataclasses._MISSING_TYPE object>,init=True,repr=True,hash=None,compare=True,metadata=mappingproxy({}),kw_only=False,_field_type=_FIELD), 'type': Field(name='type',type='builtins.type',default=<dataclasses._MISSING_TYPE object>,default_factory=<dataclasses._MISSING_TYPE object>,init=True,repr=True,hash=None,compare=True,metadata=mappingproxy({}),kw_only=False,_field_type=_FIELD)}

__dataclass_params__ = _DataclassParams(init=True,repr=False,eq=False,order=False,unsafe_hash=False,frozen=True,match_args=True,kw_only=False,slots=True,weakref_slot=False)

__delattr__(name): Implement delattr(self, name).

__eq__(other, /)[source]

Check if the path entries are equal.

Return type:: bool

__getstate__(): Helper for pickle.

__hash__()[source]

Get the hash of the path entry.

Return type:: int

__init__(entry, type, kind)

__match_args__ = ('entry', 'type', 'kind')

__post_init__()[source]

Post-initialize the path entry.

Return type:: None

__repr__()[source]

Get the representation of the path entry.

Return type:: str

__setattr__(name, value): Implement setattr(self, name, value).

__setstate__(state)

__slots__ = ('entry', 'type', 'kind')

codify(node='')[source]

Generate code for accessing the path entry.

Return type:: str

class optree.GetAttrEntry(entry, type, kind)

Bases: PyTreeEntry

A generic path entry class for nodes that access their children by __getattr__().

__annotations__ = {'__slots__': 'ClassVar[tuple[()]]', 'entry': 'str'}

__call__(obj, /)[source]

Get the child object.

Return type:: Any

__slots__: ClassVar[tuple[()]] = ()

codify(node='')[source]

Generate code for accessing the path entry.

Return type:: str

property name: str: Get the attribute name.

class optree.GetItemEntry(entry, type, kind)

Bases: PyTreeEntry

A generic path entry class for nodes that access their children by __getitem__().

__annotations__ = {'__slots__': 'ClassVar[tuple[()]]'}

__call__(obj, /)[source]

Get the child object.

Return type:: Any

__slots__: ClassVar[tuple[()]] = ()

codify(node='')[source]

Generate code for accessing the path entry.

Return type:: str

class optree.FlattenedEntry(entry, type, kind)

Bases: PyTreeEntry

A fallback path entry class for flattened objects.

__annotations__ = {'__slots__': 'ClassVar[tuple[()]]'}

__slots__: ClassVar[tuple[()]] = ()

class optree.AutoEntry(entry, type, kind)

Bases: PyTreeEntry

A generic path entry class that determines the entry type on creation automatically.

Create a new path entry.

__annotations__ = {'__slots__': 'ClassVar[tuple[()]]'}

static __new__(cls, /, entry, type, kind)[source]

Create a new path entry.

Return type:: PyTreeEntry

__slots__: ClassVar[tuple[()]] = ()

class optree.SequenceEntry(entry, type, kind)

Bases: GetItemEntry, Generic[_T_co]

A path entry class for sequences.

__annotations__ = {'__slots__': 'ClassVar[tuple[()]]', 'entry': 'int', 'type': 'builtins.type[Sequence[_T_co]]'}

__call__(obj, /)[source]

Get the child object.

Return type:: TypeVar(_T_co, covariant=True)

__orig_bases__ = (<class 'optree.GetItemEntry'>, typing.Generic[+_T_co])

__parameters__ = (+_T_co,)

__repr__()[source]

Get the representation of the path entry.

Return type:: str

__slots__: ClassVar[tuple[()]] = ()

property index: int: Get the index.

class optree.MappingEntry(entry, type, kind)

Bases: GetItemEntry, Generic[_KT_co, _VT_co]

A path entry class for mappings.

__annotations__ = {'__slots__': 'ClassVar[tuple[()]]', 'entry': '_KT_co', 'type': 'builtins.type[Mapping[_KT_co, _VT_co]]'}

__call__(obj, /)[source]

Get the child object.

Return type:: TypeVar(_VT_co, covariant=True)

__orig_bases__ = (<class 'optree.GetItemEntry'>, typing.Generic[+_KT_co, +_VT_co])

__parameters__ = (+_KT_co, +_VT_co)

__repr__()[source]

Get the representation of the path entry.

Return type:: str

__slots__: ClassVar[tuple[()]] = ()

property key: _KT_co: Get the key.

class optree.NamedTupleEntry(entry, type, kind)

Bases: SequenceEntry[_T]

A path entry class for namedtuple objects.

__annotations__ = {'__slots__': 'ClassVar[tuple[()]]', 'entry': 'int', 'kind': 'Literal[PyTreeKind.NAMEDTUPLE]', 'type': 'builtins.type[NamedTuple[_T]]'}

__orig_bases__ = (optree.SequenceEntry[~_T],)

__parameters__ = (~_T,)

__repr__()[source]

Get the representation of the path entry.

Return type:: str

__slots__: ClassVar[tuple[()]] = ()

codify(node='')[source]

Generate code for accessing the path entry.

Return type:: str

property field: str: Get the field name.

property fields: tuple[str, ...]: Get the field names.

class optree.StructSequenceEntry(entry, type, kind)

Bases: SequenceEntry[_T]

A path entry class for PyStructSequence objects.

__annotations__ = {'__slots__': 'ClassVar[tuple[()]]', 'entry': 'int', 'kind': 'Literal[PyTreeKind.STRUCTSEQUENCE]', 'type': 'builtins.type[StructSequence[_T]]'}

__orig_bases__ = (optree.SequenceEntry[~_T],)

__parameters__ = (~_T,)

__repr__()[source]

Get the representation of the path entry.

Return type:: str

__slots__: ClassVar[tuple[()]] = ()

codify(node='')[source]

Generate code for accessing the path entry.

Return type:: str

property field: str: Get the field name.

property fields: tuple[str, ...]: Get the field names.

class optree.DataclassEntry(entry, type, kind)

Bases: GetAttrEntry

A path entry class for dataclasses.

__annotations__ = {'__slots__': 'ClassVar[tuple[()]]', 'entry': 'str | int'}

__repr__()[source]

Get the representation of the path entry.

Return type:: str

__slots__: ClassVar[tuple[()]] = ()

property field: str: Get the field name.

property fields: tuple[str, ...]: Get all field names.

property init_fields: tuple[str, ...]: Get the init field names.

property name: str: Get the attribute name.

class optree.PyTreeAccessor(path: Iterable[PyTreeEntry] = ())

Bases: tuple[PyTreeEntry, …]

A path class for PyTrees.

Create a new accessor instance.

__add__(other, /)[source]

Join the accessor with another path entry or accessor.

Return type:: Self

__annotations__ = {'__slots__': 'ClassVar[tuple[()]]'}

__call__(obj, /)[source]

Get the child object.

Return type:: Any

__eq__(other, /)[source]

Check if the accessors are equal.

Return type:: bool

__getitem__(index, /)[source]

Get the child path entry or an accessor for a subpath.

Return type:: PyTreeEntry | Self

__hash__()[source]

Get the hash of the accessor.

Return type:: int

__mul__(value, /)[source]

Repeat the accessor.

Return type:: Self

static __new__(cls, /, path=())[source]

Create a new accessor instance.

Return type:: Self

__orig_bases__ = (tuple[optree.PyTreeEntry, ...],)

__repr__()[source]

Get the representation of the accessor.

Return type:: str

__rmul__(value, /)[source]

Repeat the accessor.

Return type:: Self

__slots__: ClassVar[tuple[()]] = ()

codify(root='*')[source]

Generate code for accessing the path.

Return type:: str

property path: tuple[Any, ...]: Get the path of the accessor.

optree.register_pytree_node(cls, /, flatten_func, unflatten_func, *, path_entry_type=<class 'optree.AutoEntry'>, namespace)[source]

Extend the set of types that are considered internal nodes in pytrees.

The namespace argument is used to avoid collisions that occur when different libraries register the same Python type with different behaviors. It is recommended to add a unique prefix to the namespace to avoid conflicts with other libraries. Namespaces can also be used to specify the same class in different namespaces for different use cases.

Warning

For safety reasons, a namespace must be specified while registering a custom type. It is used to isolate the behavior of flattening and unflattening a pytree node type. This is to prevent accidental collisions between different libraries that may register the same type.

Parameters:

cls (type) – A Python type to treat as an internal pytree node.
flatten_func (callable) – A function to be used during flattening, taking an instance of cls and returning a triple or optionally a pair, with (1) an iterable for the children to be flattened recursively, and (2) some hashable metadata to be stored in the treespec and to be passed to the unflatten_func, and (3) (optional) an iterable for the tree path entries to the corresponding children. If the entries are not provided or given by None, then range(len(children)) will be used.
unflatten_func (callable) – A function taking two arguments: the metadata that was returned by flatten_func and stored in the treespec, and the unflattened children. The function should return an instance of cls.
path_entry_type (type, optional) – The type of the path entry to be used in the treespec. (default: AutoEntry)
namespace (str) – A non-empty string that uniquely identifies the namespace of the type registry. This is used to isolate the registry from other modules that might register a different custom behavior for the same type.

Return type:

type[Collection[TypeVar(T)]]

Returns:

The same type as the input cls.

Raises:

TypeError – If the input type is not a class.
TypeError – If the path entry class is not a subclass of PyTreeEntry.
TypeError – If the namespace is not a string.
ValueError – If the namespace is an empty string.
ValueError – If the type is already registered in the registry.

Examples

>>> # Registry a Python type with lambda functions
>>> register_pytree_node(
...     set,
...     lambda s: (sorted(s), None, None),
...     lambda _, children: set(children),
...     namespace='set',
... )
<class 'set'>

>>> # Register a Python type into a namespace
>>> import torch
>>> register_pytree_node(
...     torch.Tensor,
...     flatten_func=lambda tensor: (
...         (tensor.cpu().detach().numpy(),),
...         {'dtype': tensor.dtype, 'device': tensor.device, 'requires_grad': tensor.requires_grad},
...     ),
...     unflatten_func=lambda metadata, children: torch.tensor(children[0], **metadata),
...     namespace='torch2numpy',
... )
<class 'torch.Tensor'>

>>>
>>> tree = {'weight': torch.ones(size=(1, 2)).cuda(), 'bias': torch.zeros(size=(2,))}
>>> tree
{'weight': tensor([[1., 1.]], device='cuda:0'), 'bias': tensor([0., 0.])}

>>> # Flatten without specifying the namespace
>>> tree_flatten(tree)  # `torch.Tensor`s are leaf nodes
([tensor([0., 0.]), tensor([[1., 1.]], device='cuda:0')], PyTreeSpec({'bias': *, 'weight': *}))

>>> # Flatten with the namespace
>>> tree_flatten(tree, namespace='torch2numpy')
(
    [array([0., 0.], dtype=float32), array([[1., 1.]], dtype=float32)],
    PyTreeSpec(
        {
            'bias': CustomTreeNode(Tensor[{'dtype': torch.float32, 'device': device(type='cpu'), 'requires_grad': False}], [*]),
            'weight': CustomTreeNode(Tensor[{'dtype': torch.float32, 'device': device(type='cuda', index=0), 'requires_grad': False}], [*])
        },
        namespace='torch2numpy'
    )
)

>>> # Register the same type with a different namespace for different behaviors
>>> def tensor2flatparam(tensor):
...     return [torch.nn.Parameter(tensor.reshape(-1))], tensor.shape, None
...
... def flatparam2tensor(metadata, children):
...     return children[0].reshape(metadata)
...
... register_pytree_node(
...     torch.Tensor,
...     flatten_func=tensor2flatparam,
...     unflatten_func=flatparam2tensor,
...     namespace='tensor2flatparam',
... )
<class 'torch.Tensor'>

>>> # Flatten with the new namespace
>>> tree_flatten(tree, namespace='tensor2flatparam')
(
    [
        Parameter containing: tensor([0., 0.], requires_grad=True),
        Parameter containing: tensor([1., 1.], device='cuda:0', requires_grad=True)
    ],
    PyTreeSpec(
        {
            'bias': CustomTreeNode(Tensor[torch.Size([2])], [*]),
            'weight': CustomTreeNode(Tensor[torch.Size([1, 2])], [*])
        },
        namespace='tensor2flatparam'
    )
)

optree.register_pytree_node_class(cls=None, /, *, path_entry_type=None, namespace=None)[source]

Extend the set of types that are considered internal nodes in pytrees.

The namespace argument is used to avoid collisions that occur when different libraries register the same Python type with different behaviors. It is recommended to add a unique prefix to the namespace to avoid conflicts with other libraries. Namespaces can also be used to specify the same class in different namespaces for different use cases.

Warning

For safety reasons, a namespace must be specified while registering a custom type. It is used to isolate the behavior of flattening and unflattening a pytree node type. This is to prevent accidental collisions between different libraries that may register the same type.

Parameters:

cls (type, optional) – A Python type to treat as an internal pytree node.
path_entry_type (type, optional) – The type of the path entry to be used in the treespec. (default: AutoEntry)
namespace (str, optional) – A non-empty string that uniquely identifies the namespace of the type registry. This is used to isolate the registry from other modules that might register a different custom behavior for the same type.

Return type:

TypeVar(CustomTreeNodeType, bound= type[CustomTreeNode]) | Callable[[TypeVar(CustomTreeNodeType, bound= type[CustomTreeNode])], TypeVar(CustomTreeNodeType, bound= type[CustomTreeNode])]

Returns:

The same type as the input cls if the argument presents. Otherwise, return a decorator function that registers the class as a pytree node.

Raises:

TypeError – If the path entry class is not a subclass of PyTreeEntry.
TypeError – If the namespace is not a string.
ValueError – If the namespace is an empty string.
ValueError – If the type is already registered in the registry.

This function is a thin wrapper around register_pytree_node(), and provides a class-oriented interface:

@register_pytree_node_class(namespace='foo')
class Special:
    TREE_PATH_ENTRY_TYPE = GetAttrEntry

    def __init__(self, x, y):
        self.x = x
        self.y = y

    def tree_flatten(self):
        return ((self.x, self.y), None, ('x', 'y'))

    @classmethod
    def tree_unflatten(cls, metadata, children):
        return cls(*children)

@register_pytree_node_class('mylist')
class MyList(UserList):
    TREE_PATH_ENTRY_TYPE = SequenceEntry

    def tree_flatten(self):
        return self.data, None, None

    @classmethod
    def tree_unflatten(cls, metadata, children):
        return cls(*children)

optree.unregister_pytree_node(cls, /, *, namespace)[source]

Remove a type from the pytree node registry.

This function is the inverse operation of function register_pytree_node().

Parameters:

cls (type) – A Python type to remove from the pytree node registry.
namespace (str) – The namespace of the pytree node registry to remove the type from.

Return type:

PyTreeNodeRegistryEntry

Returns:

The removed registry entry.

Raises:

TypeError – If the input type is not a class.
TypeError – If the namespace is not a string.
ValueError – If the namespace is an empty string.
ValueError – If the type is a built-in type that cannot be unregistered.
ValueError – If the type is not found in the registry.

Examples

>>> # Register a Python type with lambda functions
>>> register_pytree_node(
...     set,
...     lambda s: (sorted(s), None, None),
...     lambda _, children: set(children),
...     namespace='temp',
... )
<class 'set'>

>>> # Unregister the Python type
>>> unregister_pytree_node(set, namespace='temp')

optree.dict_insertion_ordered(mode, /, *, namespace)[source]

Context manager to temporarily set the dictionary sorting mode.

This context manager is used to temporarily set the dictionary sorting mode for a specific namespace. The dictionary sorting mode is used to determine whether the keys of a dictionary should be sorted or keeping the insertion order when flattening a pytree.

>>> tree = {'b': (2, [3, 4]), 'a': 1, 'c': None, 'd': 5}
>>> tree_flatten(tree)
(
    [1, 2, 3, 4, 5],
    PyTreeSpec({'a': *, 'b': (*, [*, *]), 'c': None, 'd': *})
)
>>> with dict_insertion_ordered(True, namespace='some-namespace'):
...     tree_flatten(tree, namespace='some-namespace')
(
    [2, 3, 4, 1, 5],
    PyTreeSpec({'b': (*, [*, *]), 'a': *, 'c': None, 'd': *}, namespace='some-namespace')
)

Warning

The dictionary sorting mode is a global setting and is not thread-safe. It is recommended to use this context manager in a single-threaded environment.

Parameters:

mode (bool) – The dictionary sorting mode to set.
namespace (str) – The namespace to set the dictionary sorting mode for.

Return type:

Generator[None]

class optree.PyTreeSpec

Bases: pybind11_object

Representing the structure of the pytree.

__eq__(self: optree.PyTreeSpec, other: optree.PyTreeSpec, /) → bool: Test for equality to another object.

__ge__(self: optree.PyTreeSpec, other: optree.PyTreeSpec, /) → bool: Test for this treespec is a suffix of another object.

__getstate__(self: optree.PyTreeSpec) → object

__gt__(self: optree.PyTreeSpec, other: optree.PyTreeSpec, /) → bool: Test for this treespec is a strict suffix of another object.

__hash__(self: optree.PyTreeSpec) → int: Return the hash of the treespec.

__init__(*args, **kwargs)

__le__(self: optree.PyTreeSpec, other: optree.PyTreeSpec, /) → bool: Test for this treespec is a prefix of another object.

__len__(self: optree.PyTreeSpec) → int: Number of leaves in the tree.

__lt__(self: optree.PyTreeSpec, other: optree.PyTreeSpec, /) → bool: Test for this treespec is a strict prefix of another object.

__ne__(self: optree.PyTreeSpec, other: optree.PyTreeSpec, /) → bool: Test for inequality to another object.

__pybind11_module_local_v5_gcc_libstdcpp_cxxabi1018__ = <capsule object NULL>

__setstate__(self: optree.PyTreeSpec, state: object, /) → None: Serialization support for PyTreeSpec.

accessors(self: optree.PyTreeSpec) → list[object]: Return a list of accessors to the leaves in the treespec.

broadcast_to_common_suffix(self: optree.PyTreeSpec, other: optree.PyTreeSpec, /) → optree.PyTreeSpec: Broadcast to the common suffix of this treespec and other treespec.

child(self: optree.PyTreeSpec, index: int, /) → optree.PyTreeSpec: Return the treespec for the child at the given index.

children(self: optree.PyTreeSpec) → list[optree.PyTreeSpec]: Return a list of treespecs for the children.

compose(self: optree.PyTreeSpec, inner: optree.PyTreeSpec, /) → optree.PyTreeSpec: Compose two treespecs. Constructs the inner treespec as a subtree at each leaf node.

entries(self: optree.PyTreeSpec) → list: Return a list of one-level entries to the children.

entry(self: optree.PyTreeSpec, index: int, /) → object: Return the entry at the given index.

flatten_up_to(self: optree.PyTreeSpec, tree: object, /) → list: Flatten the subtrees in tree up to the structure of this treespec and return a list of subtrees.

is_leaf(self: optree.PyTreeSpec, /, *, strict: bool = True) → bool: Test whether the treespec represents a leaf.

is_one_level(self: optree.PyTreeSpec) → bool: Test whether the treespec represents a one-level tree.

is_prefix(self: optree.PyTreeSpec, other: optree.PyTreeSpec, /, *, strict: bool = False) → bool: Test whether this treespec is a prefix of the given treespec.

is_suffix(self: optree.PyTreeSpec, other: optree.PyTreeSpec, /, *, strict: bool = False) → bool: Test whether this treespec is a suffix of the given treespec.

property kind: The kind of the root node.

property namespace: The registry namespace used to resolve the custom pytree node types.

property none_is_leaf: Whether to treat None as a leaf. If false, None is a non-leaf node with arity 0. Thus None is contained in the treespec rather than in the leaves list.

property num_children: Number of children of the root node. Note that a leaf is also a node but has no children.

property num_leaves: Number of leaves in the tree.

property num_nodes: Number of nodes in the tree. Note that a leaf is also a node but has no children.

one_level(self: optree.PyTreeSpec) → optree.PyTreeSpec | None: Return the one-level structure of the root node. Return None if the root node represents a leaf.

paths(self: optree.PyTreeSpec) → list[tuple]: Return a list of paths to the leaves of the treespec.

transform(self: optree.PyTreeSpec, /, f_node: Callable | None = None, f_leaf: Callable | None = None) → optree.PyTreeSpec: Transform the pytree structure by applying f_node(nodespec) at nodes and f_leaf(leafspec) at leaves.

traverse(self: optree.PyTreeSpec, leaves: Iterable, /, f_node: Callable | None = None, f_leaf: Callable | None = None) → object: Walk over the pytree structure, calling f_leaf(leaf) at leaves, and f_node(node) at reconstructed non-leaf nodes.

property type: The type of the root node. Return None if the root node is a leaf.

unflatten(self: optree.PyTreeSpec, leaves: Iterable, /) → object: Reconstruct a pytree from the leaves.

walk(self: optree.PyTreeSpec, leaves: Iterable, /, f_node: Callable | None = None, f_leaf: Callable | None = None) → object: Walk over the pytree structure, calling f_leaf(leaf) at leaves, and f_node(node_type, node_data, children) at non-leaf nodes.

optree.PyTreeDef: alias of PyTreeSpec

class optree.PyTreeKind(self: optree._C.PyTreeKind, value: int)

Bases: pybind11_object

The kind of a pytree node.

Members:

CUSTOM : A custom type.

LEAF : An opaque leaf node.

NONE : None.

TUPLE : A tuple.

LIST : A list.

DICT : A dict.

NAMEDTUPLE : A collections.namedtuple.

ORDEREDDICT : A collections.OrderedDict.

DEFAULTDICT : A collections.defaultdict.

DEQUE : A collections.deque.

STRUCTSEQUENCE : A PyStructSequence.

CUSTOM = <PyTreeKind.CUSTOM: 0>

DEFAULTDICT = <PyTreeKind.DEFAULTDICT: 8>

DEQUE = <PyTreeKind.DEQUE: 9>

DICT = <PyTreeKind.DICT: 5>

LEAF = <PyTreeKind.LEAF: 1>

LIST = <PyTreeKind.LIST: 4>

NAMEDTUPLE = <PyTreeKind.NAMEDTUPLE: 6>

NONE = <PyTreeKind.NONE: 2>

NUM_KINDS = 11

ORDEREDDICT = <PyTreeKind.ORDEREDDICT: 7>

STRUCTSEQUENCE = <PyTreeKind.STRUCTSEQUENCE: 10>

TUPLE = <PyTreeKind.TUPLE: 3>

__annotations__ = {}

__eq__(self: object, other: object) → bool

__getstate__(self: object) → int

__hash__(self: object) → int

__index__(self: optree._C.PyTreeKind) → int

__init__(self: optree._C.PyTreeKind, value: int) → None

__int__(self: optree._C.PyTreeKind) → int

__members__ = {'CUSTOM': <PyTreeKind.CUSTOM: 0>, 'DEFAULTDICT': <PyTreeKind.DEFAULTDICT: 8>, 'DEQUE': <PyTreeKind.DEQUE: 9>, 'DICT': <PyTreeKind.DICT: 5>, 'LEAF': <PyTreeKind.LEAF: 1>, 'LIST': <PyTreeKind.LIST: 4>, 'NAMEDTUPLE': <PyTreeKind.NAMEDTUPLE: 6>, 'NONE': <PyTreeKind.NONE: 2>, 'ORDEREDDICT': <PyTreeKind.ORDEREDDICT: 7>, 'STRUCTSEQUENCE': <PyTreeKind.STRUCTSEQUENCE: 10>, 'TUPLE': <PyTreeKind.TUPLE: 3>}

__ne__(self: object, other: object) → bool

__pybind11_module_local_v5_gcc_libstdcpp_cxxabi1018__ = <capsule object NULL>

__setstate__(self: optree._C.PyTreeKind, state: int) → None

property name

property value

class optree.PyTree[source]

Bases: Generic[T]

Generic PyTree type.

>>> import torch
>>> TensorTree = PyTree[torch.Tensor]
>>> TensorTree
typing.Union[torch.Tensor,
             tuple[ForwardRef('PyTree[torch.Tensor]'), ...],
             list[ForwardRef('PyTree[torch.Tensor]')],
             dict[typing.Any, ForwardRef('PyTree[torch.Tensor]')],
             collections.deque[ForwardRef('PyTree[torch.Tensor]')],
             optree.typing.CustomTreeNode[ForwardRef('PyTree[torch.Tensor]')]]

Prohibit instantiation.

__annotations__ = {'__instance_lock__': 'ClassVar[threading.Lock]', '__instances__': 'ClassVar[WeakKeyDictionary[TypeAliasType, tuple[type | TypeAliasType, str | None]]]', '__slots__': 'ClassVar[tuple[()]]'}

classmethod __class_getitem__(cls, item)[source]

Instantiate a PyTree type with the given type.

Return type:: TypeAliasType

__contains__(key, /)[source]

Emulate collection-like behavior.

Return type:: bool

__getattr__(name, /)[source]

Emulate dataclass-like behavior.

Return type:: TypeVar(T) | tuple[PyTree [TypeVar(T)], ...] | list[PyTree [TypeVar(T)]] | dict[Any, PyTree [TypeVar(T)]] | deque[PyTree [TypeVar(T)]] | CustomTreeNode[PyTree [TypeVar(T)]]

__getitem__(key, /)[source]

Emulate collection-like behavior.

Return type:: TypeVar(T) | tuple[PyTree [TypeVar(T)], ...] | list[PyTree [TypeVar(T)]] | dict[Any, PyTree [TypeVar(T)]] | deque[PyTree [TypeVar(T)]] | CustomTreeNode[PyTree [TypeVar(T)]]

classmethod __init_subclass__(*args, **kwargs)[source]

Prohibit subclassing.

Return type:: Never

__instance_lock__: ClassVar[allocate_lock] = <unlocked _thread.lock object>

__instances__: ClassVar[WeakKeyDictionary[TypeAliasType, tuple[type | TypeAliasType, str | None]]] = <WeakKeyDictionary>

__iter__()[source]

Emulate collection-like behavior.

Return type:: Iterator[TypeVar(T) | tuple[PyTree [TypeVar(T)], ...] | list[PyTree [TypeVar(T)]] | dict[Any, PyTree [TypeVar(T)]] | deque[PyTree [TypeVar(T)]] | CustomTreeNode[PyTree [TypeVar(T)]] | Any]

__len__()[source]

Emulate collection-like behavior.

Return type:: int

static __new__(cls, /)[source]

Prohibit instantiation.

Return type:: Never

__orig_bases__ = (typing.Generic[~T],)

__parameters__ = (~T,)

__slots__: ClassVar[tuple] = ()

count(key, /)[source]

Emulate sequence-like behavior.

Return type:: int

get(key, /, default=None)[source]

Emulate mapping-like behavior.

Return type:: TypeVar(T) | None

index(key, /)[source]

Emulate sequence-like behavior.

Return type:: int

items()[source]

Emulate mapping-like behavior.

Return type:: ItemsView[Any, TypeVar(T) | tuple[PyTree [TypeVar(T)], ...] | list[PyTree [TypeVar(T)]] | dict[Any, PyTree [TypeVar(T)]] | deque[PyTree [TypeVar(T)]] | CustomTreeNode[PyTree [TypeVar(T)]]]

keys()[source]

Emulate mapping-like behavior.

Return type:: KeysView[Any]

values()[source]

Emulate mapping-like behavior.

Return type:: ValuesView[TypeVar(T) | tuple[PyTree [TypeVar(T)], ...] | list[PyTree [TypeVar(T)]] | dict[Any, PyTree [TypeVar(T)]] | deque[PyTree [TypeVar(T)]] | CustomTreeNode[PyTree [TypeVar(T)]]]

class optree.PyTreeTypeVar(name: str, param: type | TypeAliasType)[source]

Bases: object

Type variable for PyTree.

>>> import torch
>>> TensorTree = PyTreeTypeVar('TensorTree', torch.Tensor)
>>> TensorTree
typing.Union[torch.Tensor,
             tuple[ForwardRef('TensorTree'), ...],
             list[ForwardRef('TensorTree')],
             dict[typing.Any, ForwardRef('TensorTree')],
             collections.deque[ForwardRef('TensorTree')],
             optree.typing.CustomTreeNode[ForwardRef('TensorTree')]]

Instantiate a PyTree type variable with the given name and parameter.

classmethod __init_subclass__(*args, **kwargs)[source]

Prohibit subclassing.

Return type:: Never

static __new__(cls, /, name, param)[source]

Instantiate a PyTree type variable with the given name and parameter.

Return type:: TypeAliasType

class optree.CustomTreeNode(*args, **kwargs)[source]

Bases: Protocol[T]

The abstract base class for custom pytree nodes.

__abstractmethods__ = frozenset({})

__annotations__ = {}

__init__(*args, **kwargs)

__non_callable_proto_members__ = {}

__orig_bases__ = (typing.Protocol[~T],)

__parameters__ = (~T,)

__protocol_attrs__ = {'tree_flatten', 'tree_unflatten'}

classmethod __subclasshook__(other)

Abstract classes can override this to customize issubclass().

This is invoked early on by abc.ABCMeta.__subclasscheck__(). It should return True, False or NotImplemented. If it returns NotImplemented, the normal algorithm is used. Otherwise, it overrides the normal algorithm (and the outcome is cached).

tree_flatten()[source]

Flatten the custom pytree node into children and metadata.

Return type:: tuple[Iterable[TypeVar(T)], Hashable | None] | tuple[Iterable[TypeVar(T)], Hashable | None, Iterable[Any] | None]

classmethod tree_unflatten(metadata, children, /)[source]

Unflatten the children and metadata into the custom pytree node.

Return type:: Self

class optree.FlattenFunc(*args, **kwargs)[source]

Bases: Protocol[T]

The type stub class for flatten functions.

__abstractmethods__ = frozenset({'__call__'})

__annotations__ = {}

abstractmethod __call__(container, /)[source]

Flatten the container into children and metadata.

Return type:: tuple[Iterable[TypeVar(T)], Hashable | None] | tuple[Iterable[TypeVar(T)], Hashable | None, Iterable[Any] | None]

__init__(*args, **kwargs)

__orig_bases__ = (typing.Protocol[~T],)

__parameters__ = (~T,)

__protocol_attrs__ = {'__call__'}

classmethod __subclasshook__(other)

Abstract classes can override this to customize issubclass().

This is invoked early on by abc.ABCMeta.__subclasscheck__(). It should return True, False or NotImplemented. If it returns NotImplemented, the normal algorithm is used. Otherwise, it overrides the normal algorithm (and the outcome is cached).

class optree.UnflattenFunc(*args, **kwargs)[source]

Bases: Protocol[T]

The type stub class for unflatten functions.

__abstractmethods__ = frozenset({'__call__'})

__annotations__ = {}

abstractmethod __call__(metadata, children, /)[source]

Unflatten the children and metadata back into the container.

Return type:: Collection[TypeVar(T)]

__init__(*args, **kwargs)

__orig_bases__ = (typing.Protocol[~T],)

__parameters__ = (~T,)

__protocol_attrs__ = {'__call__'}

classmethod __subclasshook__(other)

Abstract classes can override this to customize issubclass().

This is invoked early on by abc.ABCMeta.__subclasscheck__(). It should return True, False or NotImplemented. If it returns NotImplemented, the normal algorithm is used. Otherwise, it overrides the normal algorithm (and the outcome is cached).

optree.is_namedtuple(obj, /)[source]

Return whether the object is an instance of namedtuple or a subclass of namedtuple.

Return type:: bool

optree.is_namedtuple_class(cls, /)[source]

Return whether the class is a subclass of namedtuple.

Return type:: bool

optree.is_namedtuple_instance(obj, /)[source]

Return whether the object is an instance of namedtuple.

Return type:: bool

optree.namedtuple_fields(obj, /)[source]

Return the field names of a namedtuple.

Return type:: tuple[str, ...]

optree.is_structseq(obj, /)[source]

Return whether the object is an instance of PyStructSequence or a class of PyStructSequence.

Return type:: bool

optree.is_structseq_class(cls, /)[source]

Return whether the class is a class of PyStructSequence.

Return type:: bool

optree.is_structseq_instance(obj, /)[source]

Return whether the object is an instance of PyStructSequence.

Return type:: bool

optree.structseq_fields(obj, /)[source]

Return the field names of a PyStructSequence.

Return type:: tuple[str, ...]

optree.__getattr__(name, /)[source]

Get an attribute from the module.

Return type:: object