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
4f847433
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
未验证
提交
4f847433
编写于
9月 26, 2022
作者:
Z
zyfncg
提交者:
GitHub
9月 26, 2022
浏览文件
操作
浏览文件
下载
电子邮件补丁
差异文件
clear extra attrs of distribute op in opmaker (#46451)
上级
fbfd1402
变更
3
隐藏空白更改
内联
并排
Showing
3 changed file
with
10 addition
and
33 deletion
+10
-33
paddle/fluid/operators/nce_op.cc
paddle/fluid/operators/nce_op.cc
+0
-28
paddle/fluid/operators/pscore/distributed_push_sparse_op.cc
paddle/fluid/operators/pscore/distributed_push_sparse_op.cc
+0
-5
paddle/phi/api/yaml/op_compat.yaml
paddle/phi/api/yaml/op_compat.yaml
+10
-0
未找到文件。
paddle/fluid/operators/nce_op.cc
浏览文件 @
4f847433
...
@@ -207,34 +207,6 @@ class NCEOpMaker : public framework::OpProtoAndCheckerMaker {
...
@@ -207,34 +207,6 @@ class NCEOpMaker : public framework::OpProtoAndCheckerMaker {
// for parameter prefetch
// for parameter prefetch
AddAttr
<
bool
>
(
"remote_prefetch"
,
""
).
SetDefault
(
false
);
AddAttr
<
bool
>
(
"remote_prefetch"
,
""
).
SetDefault
(
false
);
AddAttr
<
int
>
(
"trainer_id"
,
"trainer id from 0 ~ worker_num."
)
.
SetDefault
(
0
)
.
AsExtra
();
AddAttr
<
std
::
vector
<
int64_t
>>
(
"height_sections"
,
"Height for each output SelectedRows."
)
.
SetDefault
(
std
::
vector
<
int64_t
>
({}))
.
AsExtra
();
AddAttr
<
std
::
vector
<
std
::
string
>>
(
"epmap"
,
"(string vector, default 127.0.0.1:6164)"
"Server endpoints in the order of input variables for mapping"
)
.
SetDefault
({})
.
AsExtra
();
AddAttr
<
std
::
vector
<
std
::
string
>>
(
"table_names"
,
"(string vector, the split table names that will be fetched from "
"parameter server)"
"in the order of input variables for mapping"
)
.
SetDefault
({})
.
AsExtra
();
AddAttr
<
std
::
vector
<
int
>>
(
"custom_neg_classes"
,
"This attribute only be used in unitest. Classes "
"in this list wiil be used as negative classes "
"for every samples. Under normal conditions, "
"user should avoid setting this attribute."
)
.
SetDefault
({})
.
AsExtra
();
AddAttr
<
bool
>
(
"is_test"
,
AddAttr
<
bool
>
(
"is_test"
,
"(bool, default false) Set to true for inference "
"(bool, default false) Set to true for inference "
"only, false for training."
)
"only, false for training."
)
...
...
paddle/fluid/operators/pscore/distributed_push_sparse_op.cc
浏览文件 @
4f847433
...
@@ -113,11 +113,6 @@ class DistributedPushSparseOpMaker : public framework::OpProtoAndCheckerMaker {
...
@@ -113,11 +113,6 @@ class DistributedPushSparseOpMaker : public framework::OpProtoAndCheckerMaker {
AddAttr
<
bool
>
(
"use_cvm_op"
,
"(boolean, default false) Use cvm op or not."
)
AddAttr
<
bool
>
(
"use_cvm_op"
,
"(boolean, default false) Use cvm op or not."
)
.
SetDefault
(
false
);
.
SetDefault
(
false
);
AddAttr
<
std
::
vector
<
int
>>
(
"slots"
,
"[slot_id1, slot_id2] Slots array of Ids."
)
.
SetDefault
({})
.
AsExtra
();
AddComment
(
R"DOC(
AddComment
(
R"DOC(
Lookup Tablel Prefetch Operator.
Lookup Tablel Prefetch Operator.
This operator is used to perform lookup on parameter W,
This operator is used to perform lookup on parameter W,
...
...
paddle/phi/api/yaml/op_compat.yaml
浏览文件 @
4f847433
...
@@ -208,6 +208,10 @@
...
@@ -208,6 +208,10 @@
outputs
:
outputs
:
out
:
Out
out
:
Out
-
op
:
distributed_push_sparse
extra
:
attrs
:
[
'
int[]
slots
=
{}'
]
-
op
:
divide (elementwise_div)
-
op
:
divide (elementwise_div)
backward
:
divide_grad (elementwise_div)
backward
:
divide_grad (elementwise_div)
extra
:
extra
:
...
@@ -485,6 +489,12 @@
...
@@ -485,6 +489,12 @@
outputs
:
outputs
:
out
:
Out
out
:
Out
-
op
:
nce
backward
:
nce_grad
extra
:
attrs
:
[
int trainer_id = 0
,
'
int64_t[]
height_sections
=
{}'
,
'
str[]
epmap
=
{}'
,
'
str[]
table_names
=
{}'
,
'
int[]
custom_neg_classes
=
{}'
]
-
op
:
nearest_interp (nearest_interp_v2)
-
op
:
nearest_interp (nearest_interp_v2)
backward
:
nearest_interp_grad (nearest_interp_v2_grad)
backward
:
nearest_interp_grad (nearest_interp_v2_grad)
extra
:
extra
:
...
...
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