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
wzpei9
Paddle
提交
2289c141
P
Paddle
项目概览
wzpei9
/
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
体验新版 GitCode,发现更多精彩内容 >>
提交
2289c141
编写于
9月 26, 2016
作者:
L
liuyuan
提交者:
qingqing01
9月 26, 2016
浏览文件
操作
浏览文件
下载
电子邮件补丁
差异文件
Refine comment for CRF related headers. (#117)
上级
332194c8
变更
2
显示空白变更内容
内联
并排
Showing
2 changed file
with
25 addition
and
25 deletion
+25
-25
paddle/gserver/layers/CRFLayer.h
paddle/gserver/layers/CRFLayer.h
+1
-1
paddle/gserver/layers/LinearChainCRF.h
paddle/gserver/layers/LinearChainCRF.h
+24
-24
未找到文件。
paddle/gserver/layers/CRFLayer.h
浏览文件 @
2289c141
...
...
@@ -25,7 +25,7 @@ namespace paddle {
/**
* A layer for calculating the cost of sequential conditional random field
* model.
* See
LinearChainCRF.h
for the detail of the CRF formulation.
* See
class LinearChainCRF
for the detail of the CRF formulation.
*/
class
CRFLayer
:
public
Layer
{
public:
...
...
paddle/gserver/layers/LinearChainCRF.h
浏览文件 @
2289c141
...
...
@@ -21,39 +21,39 @@ namespace paddle {
class
LinearChainCRF
{
public:
/*
The size of para and grad must be (numClasses + 2) * numClasses
.
The first numClasses values of para are for starting weights (a
).
The next numClasses values of para are for ending weights (b
),
The remaning values are for transition weights (w
).
The probability of a state sequence s of length L
is defined as:
P(s) = (1/Z) exp(a_{s_1} + b_{s_L}
+ \sum_{l=1}^L x_{s_l}
+ \sum_{l=2}^L w_{s_{l-1},s_l})
where Z is a normalization value so that the sum of P(s)
over all possible
sequences is 1, and x
is the input feature to the CRF.
/*
*
* The size of para and grad must be \f$(numClasses + 2) * numClasses\f$
.
* The first numClasses values of para are for starting weights (\f$a\f$
).
* The next numClasses values of para are for ending weights (\f$b\f$
),
* The remaning values are for transition weights (\f$w\f$
).
*
* The probability of a state sequence s of length \f$L\f$
is defined as:
* \f$
P(s) = (1/Z) exp(a_{s_1} + b_{s_L}
*
+ \sum_{l=1}^L x_{s_l}
* + \sum_{l=2}^L w_{s_{l-1},s_l})\f$
* where \f$Z\f$ is a normalization value so that the sum of \f$P(s)\f$
over all possible
* sequences is \f$1\f$, and \f$x\f$
is the input feature to the CRF.
*/
LinearChainCRF
(
int
numClasses
,
real
*
para
,
real
*
grad
);
/*
Calculate the negative log likelihood of s given x.
The size of x must be length * numClasses. Each consecutive numClasses
values are the features for one time step.
/*
*
*
Calculate the negative log likelihood of s given x.
*
The size of x must be length * numClasses. Each consecutive numClasses
*
values are the features for one time step.
*/
real
forward
(
real
*
x
,
int
*
s
,
int
length
);
/*
Calculate the gradient with respect to x, a, b, and w.
The gradient of x will be stored in dx.
backward() can only be called after a corresponding call to forward() with
the same x, s and length.
NOTE:
The gradient is added to dx and grad (provided at constructor).
/*
*
*
Calculate the gradient with respect to x, a, b, and w.
*
The gradient of x will be stored in dx.
*
backward() can only be called after a corresponding call to forward() with
*
the same x, s and length.
* @note
The gradient is added to dx and grad (provided at constructor).
*/
void
backward
(
real
*
x
,
real
*
dx
,
int
*
s
,
int
length
);
/*
Find the most probable sequence given x. The result will be stored in s.
/*
*
*
Find the most probable sequence given x. The result will be stored in s.
*/
void
decode
(
real
*
x
,
int
*
s
,
int
length
);
...
...
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