未验证 提交 a87d78f1 编写于 作者: L liu zhengxi 提交者: GitHub

update gather_tree doc (#30693)

* update gather_tree doc, test=document_fix

* update sample code, test=document_fix

* remove tensor type, test=document_fix
上级 fef3654b
......@@ -14984,32 +14984,30 @@ def gather_tree(ids, parents):
[9 0]]]
Args:
ids(Variable): A Tensor with shape :attr:`[length, batch_size, beam_size]`
ids(Tensor): A Tensor with shape :attr:`[length, batch_size, beam_size]`
and data type :attr:`int32` or :attr:`int64`. It contains the selected
ids of all time steps.
parents(Variable): A Tensor with the same shape and data type as :attr:`ids`,
parents(Tensor): A Tensor with the same shape and data type as :attr:`ids`,
It contains the parents corresponding to selected ids when searching
among beams.
Returns:
Variable: A Tensor with the same shape and data type as :attr:`ids`. \
A Tensor with the same shape and data type as :attr:`ids`. \
It contains the full sequences. The sequences are collected from \
:attr:`ids` by backtracing according to :attr:`parents`.
Examples:
.. code-block:: python
import paddle.fluid as fluid
import paddle
ids = paddle.to_tensor([[[2, 2], [6, 1]], [[3, 9], [6, 1]], [[0, 1], [9, 0]]])
parents = paddle.to_tensor([[[0, 0], [1, 1]], [[1, 0], [1, 0]], [[0, 0], [0, 1]]])
final_sequences = paddle.nn.functional.gather_tree(ids, parents)
# [[[2, 2], [1, 6]], [[3, 3], [6, 1]], [[0, 1], [9, 0]]]
ids = fluid.layers.data(name='ids',
shape=[5, 2, 2],
dtype='int64',
append_batch_size=False)
parents = fluid.layers.data(name='parents',
shape=[5, 2, 2],
dtype='int64',
append_batch_size=False)
final_sequences = fluid.layers.gather_tree(ids, parents)
"""
if in_dygraph_mode():
return core.ops.gather_tree(ids, parents)
......
Markdown is supported
0% .
You are about to add 0 people to the discussion. Proceed with caution.
先完成此消息的编辑!
想要评论请 注册