Skip to content
体验新版
项目
组织
正在加载...
登录
切换导航
打开侧边栏
PaddlePaddle
Paddle
提交
3a4b6cda
P
Paddle
项目概览
PaddlePaddle
/
Paddle
12 个月 前同步成功
通知
2292
Star
20931
Fork
5422
代码
文件
提交
分支
Tags
贡献者
分支图
Diff
Issue
1423
列表
看板
标记
里程碑
合并请求
543
Wiki
0
Wiki
分析
仓库
DevOps
项目成员
Pages
P
Paddle
项目概览
项目概览
详情
发布
仓库
仓库
文件
提交
分支
标签
贡献者
分支图
比较
Issue
1,423
Issue
1,423
列表
看板
标记
里程碑
合并请求
543
合并请求
543
Pages
分析
分析
仓库分析
DevOps
Wiki
0
Wiki
成员
成员
收起侧边栏
关闭侧边栏
动态
分支图
创建新Issue
提交
Issue看板
体验新版 GitCode,发现更多精彩内容 >>
未验证
提交
3a4b6cda
编写于
6月 15, 2018
作者:
T
Tao Luo
提交者:
GitHub
6月 15, 2018
浏览文件
操作
浏览文件
下载
差异文件
Merge pull request #11488 from luotao1/softmax_doc
add doc of sequence_softmax and parallelDo
上级
0ddc5d86
cff5232e
变更
3
隐藏空白更改
内联
并排
Showing
3 changed file
with
85 addition
and
3 deletion
+85
-3
paddle/fluid/operators/elementwise_mul_op.cc
paddle/fluid/operators/elementwise_mul_op.cc
+1
-1
python/paddle/fluid/layers/control_flow.py
python/paddle/fluid/layers/control_flow.py
+49
-2
python/paddle/fluid/layers/nn.py
python/paddle/fluid/layers/nn.py
+35
-0
未找到文件。
paddle/fluid/operators/elementwise_mul_op.cc
浏览文件 @
3a4b6cda
...
...
@@ -15,7 +15,7 @@ limitations under the License. */
#include "paddle/fluid/operators/elementwise_mul_op.h"
#include "paddle/fluid/operators/elementwise_op.h"
namespace
ops
=
paddle
::
operators
;
REGISTER_ELEMWISE_OP
(
elementwise_mul
,
"Mul"
,
"Out = X
\\
odot
\\
Y"
);
REGISTER_ELEMWISE_OP
(
elementwise_mul
,
"Mul"
,
"Out = X
\\
\\
odot
Y"
);
REGISTER_OP_CPU_KERNEL
(
elementwise_mul
,
ops
::
ElementwiseMulKernel
<
paddle
::
platform
::
CPUDeviceContext
,
float
>
,
...
...
python/paddle/fluid/layers/control_flow.py
浏览文件 @
3a4b6cda
...
...
@@ -234,9 +234,56 @@ class BlockGuard(object):
class
ParallelDo
(
object
):
"""
ParallelDo
class
.
ParallelDo
is used to represent multi-thread data parallel processing
.
ParallelDo class is used to create a ParallelDo.
Its vanilla implementation can be shown as the following (:math:`|` means
single thread and :math:`||||` means multiple threads)
.. code-block:: text
In the forward pass
| Split input onto different devices
| Copy parameter onto different devices
|||| Compute forward pass in parallel
| Merge output from different devices
In the backward pass
| Split output@grad onto different devices
|||| Compute backward pass in parallel
| accumulate param@grad from different devices to the first device
| Merge input@grad from different devices
| Copy param@grad to the place of parallel_do_op
Examples:
.. code-block:: python
images = fluid.layers.data(name='pixel', shape=[1, 28, 28], dtype=DTYPE)
label = fluid.layers.data(name='label', shape=[1], dtype='int64')
# ParallelDo version & Single-thread version
if thread_num > 1:
places = fluid.layers.get_places(thread_num)
pd = fluid.layers.ParallelDo(places)
with pd.do():
images = pd.read_input(images)
label = pd.read_input(label)
predict = cnn_model(images)
cost = fluid.layers.cross_entropy(input=predict, label=label)
avg_cost = fluid.layers.mean(x=cost)
pd.write_output(avg_cost)
avg_cost = pd()
avg_cost = fluid.layers.mean(avg_cost)
else:
predict = cnn_model(images)
cost = fluid.layers.cross_entropy(input=predict, label=label)
avg_cost = fluid.layers.mean(x=cost)
.. warning::
It will be soon deprecated, please use ParallelExecutor instead.
"""
def
__init__
(
self
,
places
,
use_nccl
=
False
,
name
=
None
):
...
...
python/paddle/fluid/layers/nn.py
浏览文件 @
3a4b6cda
...
...
@@ -1210,6 +1210,41 @@ def sequence_conv(input,
def
sequence_softmax
(
input
,
param_attr
=
None
,
bias_attr
=
None
,
use_cudnn
=
True
):
"""
This function computes the softmax activation among all time-steps for each
sequence. The dimension of each time-step should be 1. Thus, the shape of
input Tensor can be either :math:`[N, 1]` or :math:`[N]`, where :math:`N`
is the sum of the length of all sequences.
For i-th sequence in a mini-batch:
.. math::
Out(X[lod[i]:lod[i+1]], :) =
\\
frac{\exp(X[lod[i]:lod[i+1], :])}{\sum(\exp(X[lod[i]:lod[i+1], :]))}
For example, for a mini-batch of 3 sequences with variable-length,
each containing 2, 3, 2 time-steps, the lod of which is [0, 2, 5, 7],
then softmax will be computed among :math:`X[0:2, :]`, :math:`X[2:5, :]`,
:math:`X[5:7, :]`, and :math:`N` turns out to be 7.
Args:
input (Variable): The input variable which is a LoDTensor.
bias_attr (ParamAttr|None): attributes for bias
param_attr (ParamAttr|None): attributes for parameter
use_cudnn (bool): Use cudnn kernel or not, it is valid only when the cudnn
\
library is installed. Default: True
Returns:
Variable: output of sequence_softmax
Examples:
.. code-block:: python
x = fluid.layers.data(name='x', shape=[7, 1],
dtype='float32', lod_level=1)
x_sequence_softmax = fluid.layers.sequence_softmax(input=x)
"""
helper
=
LayerHelper
(
'sequence_softmax'
,
**
locals
())
dtype
=
helper
.
input_dtype
()
softmax_out
=
helper
.
create_tmp_variable
(
dtype
)
...
...
编辑
预览
Markdown
is supported
0%
请重试
或
添加新附件
.
添加附件
取消
You are about to add
0
people
to the discussion. Proceed with caution.
先完成此消息的编辑!
取消
想要评论请
注册
或
登录