tf.keras.tree.traverse

Traverses the given nested structure, applying the given function.

tf.keras.tree.traverse(
    func, structure, top_down=True
)

The traversal is depth-first. If top_down is True (default), parents are returned before their children (giving the option to avoid traversing into a sub-tree).

Examples:

v = []
keras.tree.traverse(v.append, [(1, 2), [3], {"a": 4}], top_down=True)
[(1, 2), [3], {'a': 4}]
v
[[(1, 2), [3], {'a': 4}], (1, 2), 1, 2, [3], 3, {'a': 4}, 4]
 
v = []
keras.tree.traverse(v.append, [(1, 2), [3], {"a": 4}], top_down=False)
[(1, 2), [3], {'a': 4}]
v
[1, 2, (1, 2), 3, [3], 4, {'a': 4}, [(1, 2), [3], {'a': 4}]]

func The function to be applied to each sub-nest of the structure.

When traversing top-down: If func(subtree) is None the traversal continues into the sub-tree. If func(subtree) is not None the traversal does not continue into the sub-tree. The sub-tree will be replaced by func(subtree) in the returned structure (to replace the sub-tree with None, use the special value _MAP_TO_NONE).

When traversing bottom-up: If func(subtree) is None the traversed sub-tree is returned unaltered. If func(subtree) is not None the sub-tree will be replaced by func(subtree) in the returned structure (to replace the sub-tree with None, use the special value _MAP_TO_NONE).

structure The structure to traverse.
top_down If True, parent structures will be visited before their children.

The structured output from the traversal.