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
机器未来
Paddle
提交
0f59d4e6
P
Paddle
项目概览
机器未来
/
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
未验证
提交
0f59d4e6
编写于
6月 28, 2021
作者:
王
王明冬
提交者:
GitHub
6月 28, 2021
浏览文件
操作
浏览文件
下载
电子邮件补丁
差异文件
add compat precondition for multihead_matmul_fuse_pass_v2,v3, test=develop (#33786)
上级
7f9b8f06
变更
6
展开全部
隐藏空白更改
内联
并排
Showing
6 changed file
with
498 addition
and
245 deletion
+498
-245
paddle/fluid/framework/ir/multihead_matmul_fuse_pass.cc
paddle/fluid/framework/ir/multihead_matmul_fuse_pass.cc
+458
-211
paddle/fluid/framework/ir/multihead_matmul_fuse_pass.h
paddle/fluid/framework/ir/multihead_matmul_fuse_pass.h
+10
-12
paddle/fluid/framework/ir/multihead_matmul_fuse_pass_tester.cc
...e/fluid/framework/ir/multihead_matmul_fuse_pass_tester.cc
+19
-19
paddle/fluid/framework/ir/pass_tester_helper.h
paddle/fluid/framework/ir/pass_tester_helper.h
+5
-1
paddle/fluid/operators/compat/matmul.pbtxt
paddle/fluid/operators/compat/matmul.pbtxt
+4
-0
paddle/fluid/operators/compat/softmax.pbtxt
paddle/fluid/operators/compat/softmax.pbtxt
+2
-2
未找到文件。
paddle/fluid/framework/ir/multihead_matmul_fuse_pass.cc
浏览文件 @
0f59d4e6
此差异已折叠。
点击以展开。
paddle/fluid/framework/ir/multihead_matmul_fuse_pass.h
浏览文件 @
0f59d4e6
...
...
@@ -18,16 +18,6 @@
#include <string>
#include "paddle/fluid/framework/ir/fuse_pass_base.h"
#include "paddle/fluid/framework/ir/graph.h"
#include "paddle/fluid/framework/ir/graph_pattern_detector.h"
namespace
paddle
{
namespace
framework
{
namespace
ir
{
class
Graph
;
}
// namespace ir
}
// namespace framework
}
// namespace paddle
namespace
paddle
{
namespace
framework
{
...
...
@@ -158,22 +148,30 @@ class MultiHeadMatmulFusePass : public FusePassBase {
class
MultiHeadMatmulV2FusePass
:
public
FusePassBase
{
public:
virtual
~
MultiHeadMatmulV2FusePass
()
{}
MultiHeadMatmulV2FusePass
();
protected:
void
ApplyImpl
(
Graph
*
graph
)
const
;
const
std
::
string
name_scope_
{
"multihead_matmul_fuse_v2"
};
private:
int
BuildFusionV2
(
Graph
*
graph
,
const
std
::
string
&
name_scope
,
Scope
*
scope
)
const
;
};
class
MultiHeadMatmulV3FusePass
:
public
FusePassBase
{
public:
virtual
~
MultiHeadMatmulV3FusePass
()
{}
MultiHeadMatmulV3FusePass
();
protected:
void
ApplyImpl
(
Graph
*
graph
)
const
;
const
std
::
string
name_scope_
{
"multihead_matmul_fuse_v3"
};
private:
int
BuildFusionV3
(
Graph
*
graph
,
const
std
::
string
&
name_scope
,
Scope
*
scope
)
const
;
};
}
// namespace ir
...
...
paddle/fluid/framework/ir/multihead_matmul_fuse_pass_tester.cc
浏览文件 @
0f59d4e6
...
...
@@ -64,7 +64,7 @@ TEST(MultiHeadMatmulFusePass, basic) {
// (transpose_qkv) reshape -> reshape_qkv
// (reshape_qkv) mul -> mul_qkv
Layers
layers
;
auto
*
x
=
layers
.
data
(
"x"
,
{
128
,
768
});
auto
*
x
=
layers
.
data
(
"x"
,
{
1
,
1
28
,
768
});
auto
out
=
layers
.
layer_norm
(
x
);
auto
*
layer_out
=
out
[
0
];
...
...
@@ -72,41 +72,41 @@ TEST(MultiHeadMatmulFusePass, basic) {
auto
*
weights_1
=
layers
.
data
(
"weights1"
,
{
768
,
768
},
true
);
auto
*
weights_2
=
layers
.
data
(
"weights2"
,
{
768
,
768
},
true
);
auto
*
mul_out_0
=
layers
.
mul
(
layer_out
,
weights_0
);
auto
*
mul_out_1
=
layers
.
mul
(
layer_out
,
weights_1
);
auto
*
mul_out_2
=
layers
.
mul
(
layer_out
,
weights_2
);
auto
*
mul_out_0
=
layers
.
mul
(
layer_out
,
weights_0
,
nullptr
,
2
);
auto
*
mul_out_1
=
layers
.
mul
(
layer_out
,
weights_1
,
nullptr
,
2
);
auto
*
mul_out_2
=
layers
.
mul
(
layer_out
,
weights_2
,
nullptr
,
2
);
auto
*
b0
=
layers
.
data
(
"bias_0"
,
{
768
},
true
);
auto
*
b1
=
layers
.
data
(
"bias_1"
,
{
768
},
true
);
auto
*
b2
=
layers
.
data
(
"bias_2"
,
{
768
},
true
);
auto
*
elementwise_out_0
=
layers
.
elementwise_add
(
mul_out_0
,
b0
);
auto
*
elementwise_out_1
=
layers
.
elementwise_add
(
mul_out_1
,
b1
);
auto
*
elementwise_out_2
=
layers
.
elementwise_add
(
mul_out_2
,
b2
);
auto
*
elementwise_out_0
=
layers
.
elementwise_add
(
mul_out_0
,
b0
,
nullptr
,
2
);
auto
*
elementwise_out_1
=
layers
.
elementwise_add
(
mul_out_1
,
b1
,
nullptr
,
2
);
auto
*
elementwise_out_2
=
layers
.
elementwise_add
(
mul_out_2
,
b2
,
nullptr
,
2
);
std
::
vector
<
int
>
shape
=
{
128
,
12
,
64
};
auto
*
reshape_0
=
layers
.
reshape2
(
elementwise_out_0
,
shape
);
auto
*
reshape_1
=
layers
.
reshape2
(
elementwise_out_1
,
shape
);
auto
*
reshape_2
=
layers
.
reshape2
(
elementwise_out_2
,
shape
);
std
::
vector
<
int
>
shape
=
{
1
,
1
28
,
12
,
64
};
auto
*
reshape_0
=
layers
.
reshape2
(
elementwise_out_0
,
shape
,
true
);
auto
*
reshape_1
=
layers
.
reshape2
(
elementwise_out_1
,
shape
,
true
);
auto
*
reshape_2
=
layers
.
reshape2
(
elementwise_out_2
,
shape
,
true
);
std
::
vector
<
int
>
axis
=
{
0
,
2
,
1
,
3
};
auto
*
transpose_0
=
layers
.
transpose2
(
reshape_0
,
axis
);
auto
*
transpose_1
=
layers
.
transpose2
(
reshape_1
,
axis
);
auto
*
transpose_2
=
layers
.
transpose2
(
reshape_2
,
axis
);
auto
*
transpose_0
=
layers
.
transpose2
(
reshape_0
,
axis
,
true
);
auto
*
transpose_1
=
layers
.
transpose2
(
reshape_1
,
axis
,
true
);
auto
*
transpose_2
=
layers
.
transpose2
(
reshape_2
,
axis
,
true
);
auto
*
scale_0
=
layers
.
scale
(
transpose_0
,
0.125
,
0
,
false
);
auto
*
matmul_qk
=
layers
.
matmul
(
scale_0
,
transpose_1
);
auto
*
matmul_qk
=
layers
.
matmul
(
scale_0
,
transpose_1
,
nullptr
,
false
,
true
);
auto
*
bqk
=
layers
.
data
(
"biasqk"
,
{
76
8
},
true
);
auto
*
bqk
=
layers
.
data
(
"biasqk"
,
{
1
,
12
,
128
,
12
8
},
true
);
auto
*
elementwise_qk
=
layers
.
elementwise_add
(
matmul_qk
,
bqk
);
auto
*
softmax_qk
=
layers
.
softmax
(
elementwise_qk
,
-
1
);
auto
*
matmul_qkv
=
layers
.
matmul
(
softmax_qk
,
transpose_2
);
auto
*
transpose_qkv
=
layers
.
transpose2
(
matmul_qkv
,
{
0
,
2
,
1
,
3
});
auto
*
reshape_qkv_out
=
layers
.
reshape2
(
transpose_qkv
,
{
1
28
,
768
}
);
auto
*
transpose_qkv
=
layers
.
transpose2
(
matmul_qkv
,
{
0
,
2
,
1
,
3
}
,
true
);
auto
*
reshape_qkv_out
=
layers
.
reshape2
(
transpose_qkv
,
{
1
,
128
,
768
},
true
);
auto
*
weights_l
=
layers
.
data
(
"weightsl"
,
{
768
,
768
},
true
);
layers
.
mul
(
reshape_qkv_out
,
weights_l
);
layers
.
mul
(
reshape_qkv_out
,
weights_l
,
nullptr
,
2
);
std
::
unique_ptr
<
ir
::
Graph
>
graph
(
new
ir
::
Graph
(
layers
.
main_program
()));
graph
->
Set
(
"__param_scope__"
,
CreateParamScope
());
...
...
paddle/fluid/framework/ir/pass_tester_helper.h
浏览文件 @
0f59d4e6
...
...
@@ -293,13 +293,17 @@ struct Layers {
return
outs
;
}
VarDesc
*
matmul
(
VarDesc
*
x
,
VarDesc
*
y
,
VarDesc
*
alpha
=
nullptr
)
{
VarDesc
*
matmul
(
VarDesc
*
x
,
VarDesc
*
y
,
VarDesc
*
alpha
=
nullptr
,
bool
transpose_x
=
false
,
bool
transpose_y
=
false
)
{
VarDesc
*
out
=
lod_tensor
(
unique_name
());
OpDesc
*
op
=
program_
.
MutableBlock
(
0
)
->
AppendOp
();
op
->
SetType
(
"matmul"
);
op
->
SetInput
(
"X"
,
{
x
->
Name
()});
op
->
SetInput
(
"Y"
,
{
y
->
Name
()});
op
->
SetOutput
(
"Out"
,
{
out
->
Name
()});
op
->
SetAttr
(
"transpose_X"
,
transpose_x
);
op
->
SetAttr
(
"transpose_Y"
,
transpose_y
);
op
->
SetAttr
(
"alpha"
,
1.0
f
);
return
out
;
}
...
...
paddle/fluid/operators/compat/matmul.pbtxt
浏览文件 @
0f59d4e6
...
...
@@ -23,6 +23,10 @@ def {
}
}
extra {
attrs {
name: "head_number"
type: INT
}
attrs {
name: "Scale_out"
type: FLOAT
...
...
paddle/fluid/operators/compat/softmax.pbtxt
浏览文件 @
0f59d4e6
...
...
@@ -10,12 +10,12 @@ def {
name: "axis"
type: INT
}
}
extra {
attrs {
name: "data_format"
type: STRING
}
}
extra {
attrs {
name: "op_role"
type: INT
...
...
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