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
Crayon鑫
Paddle
提交
e7da0940
P
Paddle
项目概览
Crayon鑫
/
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
e7da0940
编写于
8月 02, 2019
作者:
C
chengduo
提交者:
GitHub
8月 02, 2019
浏览文件
操作
浏览文件
下载
电子邮件补丁
差异文件
Disable fuse optimization option (#18924)
* Disable fuse optimization test=develop
上级
c2c876f7
变更
2
显示空白变更内容
内联
并排
Showing
2 changed file
with
8 addition
and
2 deletion
+8
-2
paddle/fluid/framework/details/build_strategy.cc
paddle/fluid/framework/details/build_strategy.cc
+7
-1
paddle/fluid/framework/details/build_strategy.h
paddle/fluid/framework/details/build_strategy.h
+1
-1
未找到文件。
paddle/fluid/framework/details/build_strategy.cc
浏览文件 @
e7da0940
...
...
@@ -77,7 +77,7 @@ class ParallelExecutorPassBuilder : public ir::PassBuilder {
// Specifies the restrictions between different pass.
if
(
strategy_
.
enable_parallel_graph_
)
{
VLOG_IF
(
3
,
strategy_
.
fuse_all_optimizer_ops_
)
<<
"Currently, fuse_all_optimizer_ops doesn't work
s
under "
<<
"Currently, fuse_all_optimizer_ops doesn't work under "
"parallel_graph."
;
strategy_
.
fuse_all_optimizer_ops_
=
false
;
}
...
...
@@ -96,6 +96,12 @@ class ParallelExecutorPassBuilder : public ir::PassBuilder {
<<
"fuse_all_optimizer_ops only work in Reducer mode."
;
strategy_
.
fuse_all_reduce_ops_
=
false
;
}
if
(
strategy_
.
async_mode_
)
{
VLOG_IF
(
3
,
strategy_
.
fuse_all_optimizer_ops_
)
<<
"Currently, fuse_all_optimizer_ops doesn't work under "
"async mode."
;
strategy_
.
fuse_all_optimizer_ops_
=
false
;
}
}
void
AppendMultiGraphOptPasses
()
{
...
...
paddle/fluid/framework/details/build_strategy.h
浏览文件 @
e7da0940
...
...
@@ -89,7 +89,7 @@ struct BuildStrategy {
bool
fuse_elewise_add_act_ops_
{
false
};
// Fuse_all_optimizer_ops and fuse_all_reduce_ops require that gradients
// should not be sparse types
bool
fuse_all_optimizer_ops_
{
tru
e
};
bool
fuse_all_optimizer_ops_
{
fals
e
};
bool
fuse_all_reduce_ops_
{
false
};
// fuse_relu_depthwise_conv can fuse the `relu ->
// depthwise_conv`
...
...
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