未验证 提交 08fb079d 编写于 作者: L lilong12 提交者: GitHub

Fix the doc for shard_index api (#29183)

* update, test=develop
上级 058f1b22
...@@ -14696,9 +14696,10 @@ def deformable_roi_pooling(input, ...@@ -14696,9 +14696,10 @@ def deformable_roi_pooling(input,
return output return output
@deprecated(since="2.0.0", update_to="paddle.shard_index")
def shard_index(input, index_num, nshards, shard_id, ignore_value=-1): def shard_index(input, index_num, nshards, shard_id, ignore_value=-1):
""" """
This operator recomputes the `input` indices according to the offset of the Recompute the `input` indices according to the offset of the
shard. The length of the indices is evenly divided into N shards, and if shard. The length of the indices is evenly divided into N shards, and if
the `shard_id` matches the shard with the input index inside, the index is the `shard_id` matches the shard with the input index inside, the index is
recomputed on the basis of the shard offset, elsewise it is set to recomputed on the basis of the shard offset, elsewise it is set to
...@@ -14711,44 +14712,27 @@ def shard_index(input, index_num, nshards, shard_id, ignore_value=-1): ...@@ -14711,44 +14712,27 @@ def shard_index(input, index_num, nshards, shard_id, ignore_value=-1):
NOTE: If the length of indices cannot be evely divided by the shard number, NOTE: If the length of indices cannot be evely divided by the shard number,
the size of the last shard will be less than the calculated `shard_size` the size of the last shard will be less than the calculated `shard_size`
Examples:
::
Input:
X.shape = [4, 1]
X.data = [[1], [6], [12], [19]]
index_num = 20
nshards = 2
ignore_value = -1
if shard_id == 0, we get:
Out.shape = [4, 1]
Out.data = [[1], [6], [-1], [-1]]
if shard_id == 1, we get:
Out.shape = [4, 1]
Out.data = [[-1], [-1], [2], [9]]
Args: Args:
- **input** (Variable): Input indices, last dimension must be 1. input (Tensor): Input indices with data type int64. It's last dimension must be 1.
- **index_num** (scalar): An integer defining the range of the index. index_num (int): An integer defining the range of the index.
- **nshards** (scalar): The number of shards nshards (int): The number of shards.
- **shard_id** (scalar): The index of the current shard shard_id (int): The index of the current shard.
- **ignore_value** (scalar): An integer value out of sharded index range ignore_value (int): An integer value out of sharded index range.
Returns: Returns:
Variable: The sharded index of input. Tensor: The sharded index of input.
Examples: Examples:
.. code-block:: python .. code-block:: python
import paddle.fluid as fluid import paddle
batch_size = 32 label = paddle.to_tensor([[16], [1]], "int64")
label = fluid.data(name="label", shape=[batch_size, 1], dtype="int64") shard_label = paddle.shard_index(input=label,
shard_label = fluid.layers.shard_index(input=label, index_num=20,
index_num=20, nshards=2,
nshards=2, shard_id=0)
shard_id=0) print(shard_label)
# [[-1], [1]]
""" """
check_variable_and_dtype(input, 'input', ['int64'], 'shard_index') check_variable_and_dtype(input, 'input', ['int64'], 'shard_index')
op_type = 'shard_index' op_type = 'shard_index'
......
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