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
wmsofts
Paddle
提交
ac663353
P
Paddle
项目概览
wmsofts
/
Paddle
与 Fork 源项目一致
Fork自
PaddlePaddle / Paddle
通知
1
Star
0
Fork
0
代码
文件
提交
分支
Tags
贡献者
分支图
Diff
Issue
0
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
0
Issue
0
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
ac663353
编写于
8月 18, 2017
作者:
D
dzhwinter
提交者:
GitHub
8月 18, 2017
浏览文件
操作
浏览文件
下载
差异文件
Merge pull request #3379 from dzhwinter/mul_op
"mul op gradient"
上级
ab270c38
b59002da
变更
5
隐藏空白更改
内联
并排
Showing
5 changed file
with
71 addition
and
21 deletion
+71
-21
paddle/operators/math/math_function.cc
paddle/operators/math/math_function.cc
+4
-4
paddle/operators/mul_op.cc
paddle/operators/mul_op.cc
+21
-4
paddle/operators/mul_op.cu
paddle/operators/mul_op.cu
+2
-0
paddle/operators/mul_op.h
paddle/operators/mul_op.h
+28
-12
python/paddle/v2/framework/tests/test_mul_op.py
python/paddle/v2/framework/tests/test_mul_op.py
+16
-1
未找到文件。
paddle/operators/math/math_function.cc
浏览文件 @
ac663353
...
...
@@ -25,8 +25,8 @@ void gemm<platform::CPUPlace, float>(const CBLAS_TRANSPOSE transA,
const
float
alpha
,
const
float
*
A
,
const
float
*
B
,
const
float
beta
,
float
*
C
,
platform
::
DeviceContext
*
context
)
{
int
lda
=
K
;
int
ldb
=
N
;
int
lda
=
(
transA
==
CblasNoTrans
)
?
K
:
M
;
int
ldb
=
(
transB
==
CblasNoTrans
)
?
N
:
K
;
int
ldc
=
N
;
cblas_sgemm
(
CblasRowMajor
,
transA
,
transB
,
M
,
N
,
K
,
alpha
,
A
,
lda
,
B
,
ldb
,
beta
,
C
,
ldc
);
...
...
@@ -40,8 +40,8 @@ void gemm<platform::CPUPlace, double>(const CBLAS_TRANSPOSE transA,
const
double
*
B
,
const
double
beta
,
double
*
C
,
platform
::
DeviceContext
*
context
)
{
int
lda
=
K
;
int
ldb
=
N
;
int
lda
=
(
transA
==
CblasNoTrans
)
?
K
:
M
;
int
ldb
=
(
transB
==
CblasNoTrans
)
?
N
:
K
;
int
ldc
=
N
;
cblas_dgemm
(
CblasRowMajor
,
transA
,
transB
,
M
,
N
,
K
,
alpha
,
A
,
lda
,
B
,
ldb
,
beta
,
C
,
ldc
);
...
...
paddle/operators/mul_op.cc
浏览文件 @
ac663353
...
...
@@ -18,6 +18,8 @@
namespace
paddle
{
namespace
operators
{
using
framework
::
Tensor
;
class
MulOp
:
public
framework
::
OperatorWithKernel
{
public:
using
framework
::
OperatorWithKernel
::
OperatorWithKernel
;
...
...
@@ -59,10 +61,23 @@ class MulOpGrad : public framework::OperatorWithKernel {
using
framework
::
OperatorWithKernel
::
OperatorWithKernel
;
protected:
void
InferShape
(
const
framework
::
InferShapeContext
&
ctx
)
const
override
{}
std
::
string
DebugString
()
const
override
{
LOG
(
INFO
)
<<
"MulGrad"
;
return
""
;
void
InferShape
(
const
framework
::
InferShapeContext
&
ctx
)
const
override
{
PADDLE_ENFORCE_NOT_NULL
(
ctx
.
InputVar
(
"X"
),
"Input(X) should not be null"
);
PADDLE_ENFORCE_NOT_NULL
(
ctx
.
InputVar
(
"Y"
),
"Input(Y) should not be null"
);
PADDLE_ENFORCE_NOT_NULL
(
ctx
.
InputVar
(
framework
::
GradVarName
(
"Out"
)),
"Input(Out@GRAD) should not be null"
);
auto
x_dims
=
ctx
.
Input
<
Tensor
>
(
"X"
)
->
dims
();
auto
y_dims
=
ctx
.
Input
<
Tensor
>
(
"Y"
)
->
dims
();
auto
out_dims
=
ctx
.
Input
<
Tensor
>
(
framework
::
GradVarName
(
"Out"
))
->
dims
();
auto
*
x_grad
=
ctx
.
Output
<
Tensor
>
(
framework
::
GradVarName
(
"X"
));
auto
*
y_grad
=
ctx
.
Output
<
Tensor
>
(
framework
::
GradVarName
(
"Y"
));
PADDLE_ENFORCE
(
x_dims
[
0
]
==
out_dims
[
0
],
"Out@GRAD M X N must equal to X dims 0, M "
);
PADDLE_ENFORCE
(
y_dims
[
1
]
==
out_dims
[
1
],
"Out@GRAD M X N must equal to Y dims 1, N "
);
x_grad
->
Resize
(
x_dims
);
y_grad
->
Resize
(
y_dims
);
}
};
...
...
@@ -72,3 +87,5 @@ class MulOpGrad : public framework::OperatorWithKernel {
namespace
ops
=
paddle
::
operators
;
REGISTER_OP
(
mul
,
ops
::
MulOp
,
ops
::
MulOpMaker
,
mul_grad
,
ops
::
MulOpGrad
);
REGISTER_OP_CPU_KERNEL
(
mul
,
ops
::
MulKernel
<
paddle
::
platform
::
CPUPlace
,
float
>
);
REGISTER_OP_CPU_KERNEL
(
mul_grad
,
ops
::
MulGradKernel
<
paddle
::
platform
::
CPUPlace
,
float
>
);
paddle/operators/mul_op.cu
浏览文件 @
ac663353
...
...
@@ -17,3 +17,5 @@
namespace
ops
=
paddle
::
operators
;
REGISTER_OP_GPU_KERNEL
(
mul
,
ops
::
MulKernel
<
paddle
::
platform
::
GPUPlace
,
float
>
);
REGISTER_OP_GPU_KERNEL
(
mul_grad
,
ops
::
MulGradKernel
<
paddle
::
platform
::
GPUPlace
,
float
>
);
paddle/operators/mul_op.h
浏览文件 @
ac663353
...
...
@@ -31,18 +31,34 @@ template <typename Place, typename T>
class
MulKernel
:
public
framework
::
OpKernel
{
public:
void
Compute
(
const
framework
::
ExecutionContext
&
context
)
const
override
{
Eigen
::
array
<
Eigen
::
IndexPair
<
Eigen
::
DenseIndex
>
,
1
>
dim_pair
=
{
{
Eigen
::
IndexPair
<
Eigen
::
DenseIndex
>
(
1
,
0
)}};
auto
*
input0
=
context
.
Input
<
Tensor
>
(
"X"
);
auto
*
input1
=
context
.
Input
<
Tensor
>
(
"Y"
);
auto
*
output
=
context
.
Output
<
Tensor
>
(
"Out"
);
output
->
mutable_data
<
T
>
(
context
.
GetPlace
());
auto
X
=
EigenMatrix
<
T
>::
From
(
*
input0
);
auto
Y
=
EigenMatrix
<
T
>::
From
(
*
input1
);
auto
Z
=
EigenMatrix
<
T
>::
From
(
*
output
);
auto
&
place
=
context
.
GetEigenDevice
<
Place
>
();
Z
.
device
(
place
)
=
X
.
contract
(
Y
,
dim_pair
);
auto
*
X
=
context
.
Input
<
Tensor
>
(
"X"
);
auto
*
Y
=
context
.
Input
<
Tensor
>
(
"Y"
);
auto
*
Z
=
context
.
Output
<
Tensor
>
(
"Out"
);
Z
->
mutable_data
<
T
>
(
context
.
GetPlace
());
auto
*
device_context
=
const_cast
<
platform
::
DeviceContext
*>
(
context
.
device_context_
);
math
::
matmul
<
Place
,
T
>
(
*
X
,
false
,
*
Y
,
false
,
1
,
Z
,
0
,
device_context
);
}
};
template
<
typename
Place
,
typename
T
>
class
MulGradKernel
:
public
framework
::
OpKernel
{
public:
void
Compute
(
const
framework
::
ExecutionContext
&
ctx
)
const
override
{
auto
*
X
=
ctx
.
Input
<
Tensor
>
(
"X"
);
auto
*
Y
=
ctx
.
Input
<
Tensor
>
(
"Y"
);
auto
*
dOut
=
ctx
.
Input
<
Tensor
>
(
framework
::
GradVarName
(
"Out"
));
auto
*
dX
=
ctx
.
Output
<
Tensor
>
(
framework
::
GradVarName
(
"X"
));
auto
*
dY
=
ctx
.
Output
<
Tensor
>
(
framework
::
GradVarName
(
"Y"
));
dX
->
mutable_data
<
T
>
(
ctx
.
GetPlace
());
dY
->
mutable_data
<
T
>
(
ctx
.
GetPlace
());
auto
*
device_context
=
const_cast
<
platform
::
DeviceContext
*>
(
ctx
.
device_context_
);
// dX = dOut * Y'. dX: M x K, dOut : M x N, Y : K x N
math
::
matmul
<
Place
,
T
>
(
*
dOut
,
false
,
*
Y
,
true
,
1
,
dX
,
0
,
device_context
);
// dY = X' * dOut. dY: K x N, dOut : M x N, X : M x K
math
::
matmul
<
Place
,
T
>
(
*
X
,
true
,
*
dOut
,
false
,
1
,
dY
,
0
,
device_context
);
}
};
...
...
python/paddle/v2/framework/tests/test_mul_op.py
浏览文件 @
ac663353
import
unittest
from
op_test_util
import
OpTestMeta
import
numpy
as
np
from
gradient_checker
import
GradientChecker
,
create_op
from
op_test_util
import
OpTestMeta
class
TestMulOp
(
unittest
.
TestCase
):
...
...
@@ -15,5 +16,19 @@ class TestMulOp(unittest.TestCase):
self
.
outputs
=
{
'Out'
:
np
.
dot
(
self
.
inputs
[
'X'
],
self
.
inputs
[
'Y'
])}
class
MulGradOpTest
(
GradientChecker
):
def
test_mul
(
self
):
op
=
create_op
(
"mul"
)
inputs
=
{
'X'
:
np
.
random
.
random
((
32
,
84
)).
astype
(
"float32"
),
'Y'
:
np
.
random
.
random
((
84
,
100
)).
astype
(
"float32"
)
}
# mul op will enlarge the relative error
self
.
check_grad
(
op
,
inputs
,
set
([
"X"
,
"Y"
]),
"Out"
,
max_relative_error
=
0.5
)
# TODO(dzh,qijun) : mulgrad test case need transpose feature of blas library
if
__name__
==
'__main__'
:
unittest
.
main
()
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