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
Crayon鑫
Paddle
提交
a398e25d
P
Paddle
项目概览
Crayon鑫
/
Paddle
与 Fork 源项目一致
Fork自
PaddlePaddle / Paddle
通知
1
Star
1
Fork
0
代码
文件
提交
分支
Tags
贡献者
分支图
Diff
Issue
1
列表
看板
标记
里程碑
合并请求
0
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
Issue
1
列表
看板
标记
里程碑
合并请求
0
合并请求
0
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
提交
a398e25d
编写于
12月 18, 2017
作者:
Y
yangyaming
浏览文件
操作
浏览文件
下载
电子邮件补丁
差异文件
Expose param_attr and bias_attr.
上级
ed56ed9f
变更
2
显示空白变更内容
内联
并排
Showing
2 changed file
with
13 addition
and
1 deletion
+13
-1
paddle/operators/lstm_unit_op.cc
paddle/operators/lstm_unit_op.cc
+4
-1
python/paddle/v2/fluid/layers/nn.py
python/paddle/v2/fluid/layers/nn.py
+9
-0
未找到文件。
paddle/operators/lstm_unit_op.cc
浏览文件 @
a398e25d
...
@@ -51,7 +51,10 @@ class LstmUnitOpMaker : public framework::OpProtoAndCheckerMaker {
...
@@ -51,7 +51,10 @@ class LstmUnitOpMaker : public framework::OpProtoAndCheckerMaker {
LstmUnitOpMaker
(
framework
::
OpProto
*
proto
,
LstmUnitOpMaker
(
framework
::
OpProto
*
proto
,
framework
::
OpAttrChecker
*
op_checker
)
framework
::
OpAttrChecker
*
op_checker
)
:
OpProtoAndCheckerMaker
(
proto
,
op_checker
)
{
:
OpProtoAndCheckerMaker
(
proto
,
op_checker
)
{
AddInput
(
"X"
,
"FC input before the non-linear activation."
);
AddInput
(
"X"
,
"Lstm unit only applies non-linear activations, please make sure"
"that linear tranformation has already been applied to `X`. "
"Linear tranformation can be applied by adding a `fc` layer"
);
AddInput
(
AddInput
(
"C_prev"
,
"C_prev"
,
"The cell state tensor of last time-step in the Lstm Unit operator."
);
"The cell state tensor of last time-step in the Lstm Unit operator."
);
...
...
python/paddle/v2/fluid/layers/nn.py
浏览文件 @
a398e25d
...
@@ -5,6 +5,7 @@ All layers just related to the neural network.
...
@@ -5,6 +5,7 @@ All layers just related to the neural network.
from
..layer_helper
import
LayerHelper
from
..layer_helper
import
LayerHelper
from
..initializer
import
Normal
,
Constant
from
..initializer
import
Normal
,
Constant
from
..framework
import
Variable
from
..framework
import
Variable
from
..param_attr
import
ParamAttr
from
tensor
import
concat
from
tensor
import
concat
__all__
=
[
__all__
=
[
...
@@ -796,6 +797,8 @@ def lstm_unit(x_t,
...
@@ -796,6 +797,8 @@ def lstm_unit(x_t,
hidden_t_prev
,
hidden_t_prev
,
cell_t_prev
,
cell_t_prev
,
forget_bias
=
0.0
,
forget_bias
=
0.0
,
param_attr
=
None
,
bias_attr
=
ParamAttr
(),
main_program
=
None
,
main_program
=
None
,
startup_program
=
None
):
startup_program
=
None
):
"""Lstm unit layer. The equation of a lstm step is:
"""Lstm unit layer. The equation of a lstm step is:
...
@@ -836,6 +839,10 @@ def lstm_unit(x_t,
...
@@ -836,6 +839,10 @@ def lstm_unit(x_t,
hidden_t_prev (Variable): The hidden value of lstm unit.
hidden_t_prev (Variable): The hidden value of lstm unit.
cell_t_prev (Variable): The cell value of lstm unit.
cell_t_prev (Variable): The cell value of lstm unit.
forget_bias (float): The forget bias of lstm unit.
forget_bias (float): The forget bias of lstm unit.
param_attr (ParamAttr): The attributes of parameter weights, used to set
initializer, name etc.
bias_attr (ParamAttr): The attributes of bias weights, used to set
initializer, name etc.
main_program (Program): The main program.
main_program (Program): The main program.
startup_program (Program): the startup program.
startup_program (Program): the startup program.
...
@@ -882,6 +889,8 @@ def lstm_unit(x_t,
...
@@ -882,6 +889,8 @@ def lstm_unit(x_t,
startup_program
=
startup_program
)
startup_program
=
startup_program
)
fc_out
=
fc
(
input
=
concat_out
,
fc_out
=
fc
(
input
=
concat_out
,
size
=
4
*
size
,
size
=
4
*
size
,
param_attr
=
param_attr
,
bias_attr
=
bias_attr
,
main_program
=
main_program
,
main_program
=
main_program
,
startup_program
=
startup_program
)
startup_program
=
startup_program
)
dtype
=
x_t
.
dtype
dtype
=
x_t
.
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