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
cd43c444
P
Paddle
项目概览
PaddlePaddle
/
Paddle
大约 1 年 前同步成功
通知
2299
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
提交
cd43c444
编写于
11月 29, 2019
作者:
J
Jacek Czaja
提交者:
Tao Luo
11月 29, 2019
浏览文件
操作
浏览文件
下载
电子邮件补丁
差异文件
[MKL-DNN] LRN and Pool2d (FWD) NHWC support (#21375)
上级
add62acf
变更
16
隐藏空白更改
内联
并排
Showing
16 changed file
with
111 addition
and
39 deletion
+111
-39
paddle/fluid/framework/data_layout_transform.cc
paddle/fluid/framework/data_layout_transform.cc
+13
-3
paddle/fluid/framework/data_layout_transform.h
paddle/fluid/framework/data_layout_transform.h
+3
-4
paddle/fluid/framework/data_transform.cc
paddle/fluid/framework/data_transform.cc
+10
-1
paddle/fluid/framework/executor.cc
paddle/fluid/framework/executor.cc
+1
-0
paddle/fluid/framework/operator.h
paddle/fluid/framework/operator.h
+5
-0
paddle/fluid/operators/controlflow/fetch_op.cc
paddle/fluid/operators/controlflow/fetch_op.cc
+6
-2
paddle/fluid/operators/lrn_op.cc
paddle/fluid/operators/lrn_op.cc
+22
-6
paddle/fluid/operators/mkldnn/dequantize_mkldnn_op.cc
paddle/fluid/operators/mkldnn/dequantize_mkldnn_op.cc
+1
-0
paddle/fluid/operators/mkldnn/lrn_mkldnn_op.cc
paddle/fluid/operators/mkldnn/lrn_mkldnn_op.cc
+2
-0
paddle/fluid/operators/pool_op.cc
paddle/fluid/operators/pool_op.cc
+26
-7
paddle/fluid/operators/pool_op.h
paddle/fluid/operators/pool_op.h
+4
-0
paddle/fluid/platform/device_context.cc
paddle/fluid/platform/device_context.cc
+12
-0
paddle/fluid/platform/device_context.h
paddle/fluid/platform/device_context.h
+3
-0
paddle/fluid/platform/mkldnn_reuse.h
paddle/fluid/platform/mkldnn_reuse.h
+1
-1
python/paddle/fluid/tests/unittests/mkldnn/test_lrn_mkldnn_op.py
...paddle/fluid/tests/unittests/mkldnn/test_lrn_mkldnn_op.py
+1
-6
python/paddle/fluid/tests/unittests/mkldnn/test_pool2d_mkldnn_op.py
...dle/fluid/tests/unittests/mkldnn/test_pool2d_mkldnn_op.py
+1
-9
未找到文件。
paddle/fluid/framework/data_layout_transform.cc
浏览文件 @
cd43c444
...
...
@@ -127,13 +127,17 @@ void TransDataLayoutFromMKLDNN(const OpKernelType& kernel_type_for_var,
"TransDataLayoutFromMKLDNN only supports transform from MKLDNN to "
"non-MKLDNN"
);
innerTransDataLayoutFromMKLDNN
(
in_layout
,
out_layout
,
in
,
out
,
place
);
#ifdef PADDLE_WITH_MKLDNN
innerTransDataLayoutFromMKLDNN
(
in_layout
,
paddle
::
platform
::
get_cur_paddle_data_layout
(),
in
,
out
,
place
);
#endif
}
#ifdef PADDLE_WITH_MKLDNN
void
innerTransDataLayoutFromMKLDNN
(
DataLayout
in_layout
,
DataLayout
out_layout
,
const
Tensor
&
in
,
Tensor
*
out
,
platform
::
Place
place
)
{
#ifdef PADDLE_WITH_MKLDNN
PADDLE_ENFORCE_NE
(
in
.
format
(),
MKLDNNMemoryFormat
::
format_undef
,
platform
::
errors
::
InvalidArgument
(
"Input tensor format is invalid. Input tensor should "
...
...
@@ -185,11 +189,17 @@ void innerTransDataLayoutFromMKLDNN(DataLayout in_layout, DataLayout out_layout,
}
else
{
out
->
ShareDataWith
(
in
);
}
// For exepected NHWC data format we need to reshape the Output tensor
// As MKL-DNN description was in NCHW and paddle is expecting NHWC
if
(
out_layout
==
DataLayout
::
kNHWC
)
{
std
::
rotate
(
out_tz
.
begin
()
+
1
,
out_tz
.
begin
()
+
2
,
out_tz
.
end
());
out
->
Resize
(
framework
::
make_ddim
(
out_tz
));
}
out
->
set_layout
(
out_layout
);
// reset format since the out tensor will be feed to non-MKLDNN OPkernel
out
->
set_format
(
MKLDNNMemoryFormat
::
format_undef
);
#endif
}
#endif
}
// namespace framework
}
// namespace paddle
paddle/fluid/framework/data_layout_transform.h
浏览文件 @
cd43c444
...
...
@@ -66,16 +66,15 @@ inline MKLDNNDataType ToMKLDNNDataType(proto::VarType::Type type) {
return
MKLDNNDataType
::
data_undef
;
}
void
innerTransDataLayoutFromMKLDNN
(
DataLayout
in_layout
,
DataLayout
out_layout
,
const
Tensor
&
in
,
Tensor
*
out
,
platform
::
Place
place
);
#endif
void
TransDataLayoutFromMKLDNN
(
const
OpKernelType
&
kernel_type_for_var
,
const
OpKernelType
&
expected_kernel_type
,
const
Tensor
&
in
,
Tensor
*
out
);
void
innerTransDataLayoutFromMKLDNN
(
DataLayout
in_layout
,
DataLayout
out_layout
,
const
Tensor
&
in
,
Tensor
*
out
,
platform
::
Place
place
);
std
::
vector
<
int
>
GetAxis
(
const
DataLayout
&
from
,
const
DataLayout
&
to
);
void
TransDataLayout
(
const
OpKernelType
&
kernel_type_for_var
,
...
...
paddle/fluid/framework/data_transform.cc
浏览文件 @
cd43c444
...
...
@@ -19,6 +19,7 @@ limitations under the License. */
#include "paddle/fluid/framework/data_type_transform.h"
#ifdef PADDLE_WITH_MKLDNN
#include <algorithm>
#include "paddle/fluid/platform/mkldnn_helper.h"
#endif
...
...
@@ -54,8 +55,16 @@ void TransformData(const OpKernelType &expected_kernel_type,
auto
out_format
=
platform
::
MKLDNNFormatForSize
(
in
.
dims
().
size
(),
ToMKLDNNFormat
(
lin
));
out
.
ShareDataWith
(
input_tensor
);
// For NHWC data we need reshape of tensors as MKL-DNN
// is expecting NHWC dims description order
if
(
lin
==
DataLayout
::
kNHWC
)
{
auto
nchw_dims
=
paddle
::
framework
::
vectorize
<
int
>
(
out
.
dims
());
std
::
rotate
(
nchw_dims
.
begin
()
+
1
,
nchw_dims
.
end
()
-
1
,
nchw_dims
.
end
());
out
.
Resize
(
framework
::
make_ddim
(
nchw_dims
));
paddle
::
platform
::
set_cur_paddle_data_layout
(
lin
);
}
out
.
set_layout
(
DataLayout
::
kMKLDNN
);
out
.
set_format
(
out_format
);
#endif
...
...
paddle/fluid/framework/executor.cc
浏览文件 @
cd43c444
...
...
@@ -103,6 +103,7 @@ Executor::~Executor() {
platform
::
MKLDNNDeviceContext
*
dev_ctx
=
(
platform
::
MKLDNNDeviceContext
*
)
pool
.
Get
(
place_
);
dev_ctx
->
ResetBlobMap
();
platform
::
set_cur_paddle_data_layout
(
paddle
::
framework
::
DataLayout
::
kNCHW
);
}
#endif
}
...
...
paddle/fluid/framework/operator.h
浏览文件 @
cd43c444
...
...
@@ -470,6 +470,11 @@ class OperatorWithKernel : public OperatorBase {
return
g_all_op_kernels
;
}
bool
IsMKLDNNType
()
const
{
return
((
this
->
kernel_type_
)
&&
(
this
->
kernel_type_
->
data_layout_
==
framework
::
DataLayout
::
kMKLDNN
));
}
bool
SupportGPU
()
const
override
{
auto
&
op_kernels
=
OperatorWithKernel
::
AllOpKernels
().
at
(
type_
);
return
std
::
any_of
(
op_kernels
.
begin
(),
op_kernels
.
end
(),
...
...
paddle/fluid/operators/controlflow/fetch_op.cc
浏览文件 @
cd43c444
...
...
@@ -56,16 +56,20 @@ class FetchOp : public framework::OperatorBase {
// FIXME(yuyang18): Should we assume the fetch operator always generate
// CPU outputs?
if
(
src_item
.
IsInitialized
()
&&
src_item
.
numel
()
>
0
)
{
#ifdef PADDLE_WITH_MKLDNN
// Conversion from MKL-DNN to Paddle
if
(
src_item
.
layout
()
==
framework
::
DataLayout
::
kMKLDNN
)
{
framework
::
Tensor
out
;
framework
::
innerTransDataLayoutFromMKLDNN
(
src_item
.
layout
(),
framework
::
DataLayout
::
kNCHW
,
src_item
,
&
out
,
platform
::
CPUPlace
());
src_item
.
layout
(),
paddle
::
platform
::
get_cur_paddle_data_layout
()
,
src_item
,
&
out
,
platform
::
CPUPlace
());
TensorCopySync
(
out
,
platform
::
CPUPlace
(),
&
dst_item
);
}
else
{
TensorCopySync
(
src_item
,
platform
::
CPUPlace
(),
&
dst_item
);
}
#else
TensorCopySync
(
src_item
,
platform
::
CPUPlace
(),
&
dst_item
);
#endif
}
else
{
// Not copy, if the src tensor is empty.
dst_item
.
clear
();
...
...
paddle/fluid/operators/lrn_op.cc
浏览文件 @
cd43c444
...
...
@@ -193,12 +193,6 @@ class LRNOp : public framework::OperatorWithKernel {
#ifdef PADDLE_WITH_MKLDNN
if
(
library_
==
framework
::
LibraryType
::
kPlain
&&
platform
::
CanMKLDNNBeUsed
(
ctx
))
{
// TODO(jczaja): Add support for NHWC
const
std
::
string
data_format
=
ctx
.
Attr
<
std
::
string
>
(
"data_format"
);
PADDLE_ENFORCE_NE
(
data_format
,
"NHWC"
,
platform
::
errors
::
Unimplemented
(
"LRN MKLDNN does not support NHWC data format yet"
));
library_
=
framework
::
LibraryType
::
kMKLDNN
;
layout_
=
framework
::
DataLayout
::
kMKLDNN
;
}
...
...
@@ -207,6 +201,28 @@ class LRNOp : public framework::OperatorWithKernel {
OperatorWithKernel
::
IndicateVarDataType
(
ctx
,
"X"
),
ctx
.
GetPlace
(),
layout_
,
library_
);
}
framework
::
OpKernelType
GetKernelTypeForVar
(
const
std
::
string
&
var_name
,
const
Tensor
&
tensor
,
const
framework
::
OpKernelType
&
expected_kernel_type
)
const
override
{
#ifdef PADDLE_WITH_MKLDNN
if
((
expected_kernel_type
.
data_layout_
==
framework
::
DataLayout
::
kMKLDNN
)
&&
(
tensor
.
layout
()
!=
framework
::
DataLayout
::
kMKLDNN
))
{
auto
attrs
=
Attrs
();
auto
ar
=
paddle
::
framework
::
AttrReader
(
attrs
);
const
std
::
string
data_format
=
ar
.
Get
<
std
::
string
>
(
"data_format"
);
auto
dl
=
framework
::
StringToDataLayout
(
data_format
);
// Some models may have intentionally set "AnyLayout" for pool
// op. Treat this as NCHW (default data_format value)
if
(
dl
!=
framework
::
DataLayout
::
kAnyLayout
)
{
return
framework
::
OpKernelType
(
expected_kernel_type
.
data_type_
,
tensor
.
place
(),
dl
);
}
}
#endif
return
framework
::
OpKernelType
(
expected_kernel_type
.
data_type_
,
tensor
.
place
(),
tensor
.
layout
());
}
};
template
<
typename
T
>
...
...
paddle/fluid/operators/mkldnn/dequantize_mkldnn_op.cc
浏览文件 @
cd43c444
...
...
@@ -102,6 +102,7 @@ class DeQuantOpKernel : public framework::OpKernel<T> {
pipeline
.
push_back
(
*
reorder_p
);
stream
(
stream
::
kind
::
eager
).
submit
(
pipeline
).
wait
();
output
->
set_layout
(
DataLayout
::
kMKLDNN
);
output
->
set_format
(
GetMKLDNNFormat
(
*
dst_memory
));
}
};
...
...
paddle/fluid/operators/mkldnn/lrn_mkldnn_op.cc
浏览文件 @
cd43c444
...
...
@@ -62,6 +62,8 @@ class LRNMKLDNNOpKernel : public paddle::framework::OpKernel<T> {
std
::
shared_ptr
<
mkldnn
::
lrn_forward
>
lrn_p
;
if
(
is_test
==
false
)
{
workspace_memory
=
handler
.
AcquireWorkspaceMemory
(
mid
);
mid
->
set_layout
(
framework
::
DataLayout
::
kMKLDNN
);
mid
->
set_format
(
platform
::
GetMKLDNNFormat
(
*
workspace_memory
));
lrn_p
=
handler
.
AcquireForwardPrimitive
(
*
src_memory
,
*
workspace_memory
,
*
dst_memory
);
}
else
{
...
...
paddle/fluid/operators/pool_op.cc
浏览文件 @
cd43c444
...
...
@@ -88,7 +88,10 @@ void PoolOp::InferShape(framework::InferShapeContext* ctx) const {
ksize
.
size
(),
strides
.
size
(),
framework
::
make_ddim
(
ksize
),
framework
::
make_ddim
(
strides
));
const
bool
channel_last
=
(
data_format
==
"NHWC"
||
data_format
==
"NDHWC"
);
// MKL-DNN Kernels are using NCHW order of dims description
// so we ignore data_format consideration for MKL-DNN kernel
const
bool
channel_last
=
(
this
->
IsMKLDNNType
()
==
false
)
&&
(
data_format
==
"NHWC"
||
data_format
==
"NDHWC"
);
// update paddings if "SAME" or global_pooling
framework
::
DDim
data_dims
;
...
...
@@ -146,12 +149,6 @@ framework::OpKernelType PoolOp::GetExpectedKernelType(
#ifdef PADDLE_WITH_MKLDNN
if
(
library_
==
framework
::
LibraryType
::
kPlain
&&
platform
::
CanMKLDNNBeUsed
(
ctx
))
{
// TODO(jczaja): Add support for NHWC
const
std
::
string
data_format
=
ctx
.
Attr
<
std
::
string
>
(
"data_format"
);
PADDLE_ENFORCE_NE
(
data_format
,
"NHWC"
,
platform
::
errors
::
Unimplemented
(
"Pool MKLDNN grad does not support NHWC data format yet"
));
library_
=
framework
::
LibraryType
::
kMKLDNN
;
layout_
=
framework
::
DataLayout
::
kMKLDNN
;
}
...
...
@@ -162,6 +159,28 @@ framework::OpKernelType PoolOp::GetExpectedKernelType(
layout_
,
library_
);
}
framework
::
OpKernelType
PoolOp
::
GetKernelTypeForVar
(
const
std
::
string
&
var_name
,
const
Tensor
&
tensor
,
const
framework
::
OpKernelType
&
expected_kernel_type
)
const
{
#ifdef PADDLE_WITH_MKLDNN
if
((
expected_kernel_type
.
data_layout_
==
framework
::
DataLayout
::
kMKLDNN
)
&&
(
tensor
.
layout
()
!=
framework
::
DataLayout
::
kMKLDNN
))
{
auto
attrs
=
Attrs
();
auto
ar
=
paddle
::
framework
::
AttrReader
(
attrs
);
const
std
::
string
data_format
=
ar
.
Get
<
std
::
string
>
(
"data_format"
);
auto
dl
=
framework
::
StringToDataLayout
(
data_format
);
// Some models may have intentionally set "AnyLayout" for pool
// op. Treat this as NCHW (default data_format value)
if
(
dl
!=
framework
::
DataLayout
::
kAnyLayout
)
{
return
framework
::
OpKernelType
(
expected_kernel_type
.
data_type_
,
tensor
.
place
(),
dl
);
}
}
#endif
return
framework
::
OpKernelType
(
expected_kernel_type
.
data_type_
,
tensor
.
place
(),
tensor
.
layout
());
}
void
PoolOpGrad
::
InferShape
(
framework
::
InferShapeContext
*
ctx
)
const
{
PADDLE_ENFORCE_EQ
(
ctx
->
HasInput
(
"X"
),
true
,
"Input(X) must not be null."
);
PADDLE_ENFORCE_EQ
(
ctx
->
HasOutput
(
framework
::
GradVarName
(
"X"
)),
true
,
...
...
paddle/fluid/operators/pool_op.h
浏览文件 @
cd43c444
...
...
@@ -35,6 +35,10 @@ class PoolOp : public framework::OperatorWithKernel {
protected:
framework
::
OpKernelType
GetExpectedKernelType
(
const
framework
::
ExecutionContext
&
ctx
)
const
override
;
framework
::
OpKernelType
GetKernelTypeForVar
(
const
std
::
string
&
var_name
,
const
Tensor
&
tensor
,
const
framework
::
OpKernelType
&
expected_kernel_type
)
const
;
};
class
PoolOpGrad
:
public
framework
::
OperatorWithKernel
{
...
...
paddle/fluid/platform/device_context.cc
浏览文件 @
cd43c444
...
...
@@ -397,6 +397,10 @@ thread_local std::string cur_input_shape_str = "";
// the cache capacity of different input shapes for MKLDNN.
// Default 1 means fixed input shape, not dynamic shape.
thread_local
int
cur_input_shape_cache_capacity
=
1
;
// Recently registered data_format. This is needed to
// know for converting MKL-DNN Tensor to non MKL-DNN
thread_local
paddle
::
framework
::
DataLayout
cur_paddle_data_layout
=
paddle
::
framework
::
DataLayout
::
kNCHW
;
}
// namespace
void
set_cur_mkldnn_session_id
(
size_t
sid
)
{
cur_mkldnn_session_id
=
sid
;
}
...
...
@@ -408,6 +412,14 @@ void set_cur_input_shape_cache_capacity(int input_shape_cache_capacity) {
cur_input_shape_cache_capacity
=
input_shape_cache_capacity
;
}
void
set_cur_paddle_data_layout
(
framework
::
DataLayout
dl
)
{
cur_paddle_data_layout
=
dl
;
}
framework
::
DataLayout
get_cur_paddle_data_layout
(
void
)
{
return
cur_paddle_data_layout
;
}
void
MKLDNNDeviceContext
::
ResetBlobMap
()
const
{
p_blobmap_
->
clear
();
}
size_t
MKLDNNDeviceContext
::
GetShapeBlobSize
()
const
{
...
...
paddle/fluid/platform/device_context.h
浏览文件 @
cd43c444
...
...
@@ -30,6 +30,7 @@ limitations under the License. */
#ifdef PADDLE_WITH_MKLDNN
#include "mkldnn.hpp"
#include "paddle/fluid/framework/data_layout.h"
#endif
#include <map>
...
...
@@ -290,6 +291,8 @@ void set_cur_mkldnn_session_id(size_t);
size_t
get_cur_mkldnn_session_id
(
void
);
void
set_cur_input_shape_str
(
std
::
string
input_shape_str
);
void
set_cur_input_shape_cache_capacity
(
int
input_shape_cache_capacity
);
void
set_cur_paddle_data_layout
(
framework
::
DataLayout
);
framework
::
DataLayout
get_cur_paddle_data_layout
(
void
);
class
MKLDNNDeviceContext
:
public
CPUDeviceContext
{
public:
...
...
paddle/fluid/platform/mkldnn_reuse.h
浏览文件 @
cd43c444
...
...
@@ -502,7 +502,7 @@ class LRNMKLDNNHandler
std
::
shared_ptr
<
mkldnn
::
memory
>
AcquireWorkspaceMemory
(
framework
::
Tensor
*
workspace
)
{
T
*
ptr
=
workspace
->
mutable_data
<
T
>
(
this
->
place_
,
this
->
fwd_pd_
->
dst
_primitive_desc
().
get_size
());
this
->
place_
,
this
->
fwd_pd_
->
workspace
_primitive_desc
().
get_size
());
return
this
->
AcquireMemoryFromPrimitive
(
this
->
fwd_pd_
->
workspace_primitive_desc
(),
ptr
,
"@wrk_mem_p"
);
}
...
...
python/paddle/fluid/tests/unittests/mkldnn/test_lrn_mkldnn_op.py
浏览文件 @
cd43c444
...
...
@@ -55,16 +55,11 @@ class TestLRNMKLDNNOpWithIsTest(TestLRNMKLDNNOp):
self
.
assertRaises
(
AttributeError
,
check_raise_is_test
)
# TODO(jczaja): Once mkl-dnn integration support NHWC input
# then those tests should be changed to actual functional positive tests
class
TestLRNMKLDNNOpNHWC
(
TestLRNMKLDNNOp
):
def
init_test_case
(
self
):
self
.
data_format
=
'NHWC'
def
test_check_output
(
self
):
pass
# Grad tests both FWD and BWD ops kernels creation
#TODO(jczaja): Add grad support
def
test_check_grad_normal
(
self
):
with
self
.
assertRaises
(
fluid
.
core_avx
.
EnforceNotMet
):
self
.
check_grad
([
'X'
],
'Out'
,
max_relative_error
=
0.01
)
...
...
python/paddle/fluid/tests/unittests/mkldnn/test_pool2d_mkldnn_op.py
浏览文件 @
cd43c444
...
...
@@ -141,9 +141,6 @@ class TestAsymPadValid(TestAsymPad):
self
.
padding_algorithm
=
"VALID"
# Designed to Fail
# TODO(jczaja): Once mkl-dnn integration support NHWC input
# then those tests should be changed to actual functional positive tests
class
TestAsymPadValidNHWC
(
TestAsymPadValid
):
def
init_data_format
(
self
):
self
.
data_format
=
"NHWC"
...
...
@@ -151,12 +148,7 @@ class TestAsymPadValidNHWC(TestAsymPadValid):
def
init_shape
(
self
):
self
.
shape
=
[
2
,
7
,
7
,
3
]
def
test_check_output
(
self
):
pass
# Grad tests both FWD and BWD ops kernels creation
# GetExpectedKernelType should throw an exception on lack of support
# to NHWC inputs in pool mkldnn kernel
#TODO(jczaja): Add Grad NHWC support
def
test_check_grad
(
self
):
with
self
.
assertRaises
(
fluid
.
core_avx
.
EnforceNotMet
):
super
(
TestAsymPadValidNHWC
,
self
).
test_check_grad
()
...
...
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