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
BaiXuePrincess
Paddle
提交
37b33973
P
Paddle
项目概览
BaiXuePrincess
/
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
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
提交
37b33973
编写于
2月 10, 2023
作者:
C
cyber-pioneer
浏览文件
操作
浏览文件
下载
电子邮件补丁
差异文件
fix composite check output
上级
2590527f
变更
3
隐藏空白更改
内联
并排
Showing
3 changed file
with
12 addition
and
1 deletion
+12
-1
paddle/phi/api/yaml/op_compat.yaml
paddle/phi/api/yaml/op_compat.yaml
+7
-0
python/paddle/incubate/autograd/generate_op_map.py
python/paddle/incubate/autograd/generate_op_map.py
+1
-1
python/paddle/incubate/autograd/primx.py
python/paddle/incubate/autograd/primx.py
+4
-0
未找到文件。
paddle/phi/api/yaml/op_compat.yaml
浏览文件 @
37b33973
...
@@ -145,6 +145,13 @@
...
@@ -145,6 +145,13 @@
variance
:
Variance
variance
:
Variance
scale
:
Scale
scale
:
Scale
bias
:
Bias
bias
:
Bias
outputs
:
out
:
Y
mean_out
:
MeanOut
variance_out
:
VarianceOut
saved_mean
:
SavedMean
saved_variance
:
SavedVariance
reserve_space
:
ReserveSpace
extra
:
extra
:
attrs
:
[
bool use_mkldnn = false
,
bool fuse_with_relu = false
]
attrs
:
[
bool use_mkldnn = false
,
bool fuse_with_relu = false
]
...
...
python/paddle/incubate/autograd/generate_op_map.py
浏览文件 @
37b33973
...
@@ -84,7 +84,7 @@ def generate_code(
...
@@ -84,7 +84,7 @@ def generate_code(
else
:
else
:
op_name
=
key
op_name
=
key
map_dct
[
op_name
]
=
{
"phi_name"
:
op_name
}
map_dct
[
op_name
]
=
{
"phi_name"
:
op_name
}
for
element
in
[
"inputs"
,
"attrs"
]:
for
element
in
[
"inputs"
,
"
outputs"
,
"
attrs"
]:
if
element
in
item
.
keys
():
if
element
in
item
.
keys
():
map_dct
[
op_name
][
element
]
=
item
[
element
]
map_dct
[
op_name
][
element
]
=
item
[
element
]
for
element
in
[
"scalar"
,
"int_array"
]:
for
element
in
[
"scalar"
,
"int_array"
]:
...
...
python/paddle/incubate/autograd/primx.py
浏览文件 @
37b33973
...
@@ -609,6 +609,10 @@ def _lower_composite(block, blacklist=[]):
...
@@ -609,6 +609,10 @@ def _lower_composite(block, blacklist=[]):
expand_nested_list
(
as_tensors
(
lower_fn
(
op
,
*
input_args
))),
expand_nested_list
(
as_tensors
(
lower_fn
(
op
,
*
input_args
))),
):
):
if
new_out
is
not
None
:
if
new_out
is
not
None
:
assert
orig_out
.
shape
==
new_out
.
shape
,
(
f
'when replace origin op with composite rule, origin out shape should be equal to new out shape, '
f
'but orig_out.shape=
{
orig_out
.
shape
}
and new_out.shape=
{
new_out
.
shape
}
'
)
assert
not
(
orig_out
is
None
)
^
(
assert
not
(
orig_out
is
None
)
^
(
new_out
is
None
new_out
is
None
),
"orig_out and new_out should match."
),
"orig_out and new_out should match."
...
...
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