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
OPTHREE
Paddle
提交
e2ca4240
P
Paddle
项目概览
OPTHREE
/
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
体验新版 GitCode,发现更多精彩内容 >>
提交
e2ca4240
编写于
4月 25, 2018
作者:
F
fengjiayi
浏览文件
操作
浏览文件
下载
电子邮件补丁
差异文件
Replace Channel in DoubleBufferReader with BlockingQueue
上级
a7866113
变更
1
隐藏空白更改
内联
并排
Showing
1 changed file
with
6 addition
and
9 deletion
+6
-9
paddle/fluid/operators/reader/create_double_buffer_reader_op.cc
.../fluid/operators/reader/create_double_buffer_reader_op.cc
+6
-9
未找到文件。
paddle/fluid/operators/reader/create_double_buffer_reader_op.cc
浏览文件 @
e2ca4240
...
...
@@ -14,7 +14,7 @@
#include <thread> // NOLINT
#include "paddle/fluid/
framework/channel
.h"
#include "paddle/fluid/
operators/reader/blocking_queue
.h"
#include "paddle/fluid/operators/reader/reader_op_registry.h"
namespace
paddle
{
...
...
@@ -23,13 +23,13 @@ namespace reader {
// 'Double buffer' means we shall maintain two batches of input data at the same
// time. So the kCacheSize shoul be at least 2.
static
constexpr
size_t
kCacheSize
=
2
;
static
constexpr
size_t
kCacheSize
=
3
;
// There will be two bacthes out of the channel during training:
// 1. the one waiting to be sent to the channel
// 2. the one just be received from the channel, which is also being used by
// subsequent operators.
// So the channel size should be kChacheSize - 2
static
constexpr
size_t
kChannelSize
=
0
;
// kCacheSize - 2
static
constexpr
size_t
kChannelSize
=
1
;
// kCacheSize - 2
class
DoubleBufferReader
:
public
framework
::
DecoratedReader
{
public:
...
...
@@ -58,7 +58,7 @@ class DoubleBufferReader : public framework::DecoratedReader {
bool
HasNext
()
const
;
void
StartPrefetcher
()
{
channel_
=
framework
::
MakeChannel
<
size_t
>
(
kChannelSize
);
channel_
=
new
reader
::
BlockingQueue
<
size_t
>
(
kChannelSize
);
prefetcher_
=
std
::
thread
([
this
]
{
PrefetchThreadFunc
();
});
}
...
...
@@ -74,7 +74,7 @@ class DoubleBufferReader : public framework::DecoratedReader {
void
PrefetchThreadFunc
();
std
::
thread
prefetcher_
;
framework
::
Channel
<
size_t
>*
channel_
;
reader
::
BlockingQueue
<
size_t
>*
channel_
;
platform
::
Place
place_
;
std
::
vector
<
std
::
vector
<
framework
::
LoDTensor
>>
cpu_tensor_cache_
;
std
::
vector
<
std
::
vector
<
framework
::
LoDTensor
>>
gpu_tensor_cache_
;
...
...
@@ -185,10 +185,7 @@ void DoubleBufferReader::PrefetchThreadFunc() {
gpu_batch
[
i
].
set_lod
(
cpu_batch
[
i
].
lod
());
}
}
try
{
size_t
tmp
=
cached_tensor_id
;
channel_
->
Send
(
&
tmp
);
}
catch
(
paddle
::
platform
::
EnforceNotMet
e
)
{
if
(
!
channel_
->
Send
(
cached_tensor_id
))
{
VLOG
(
5
)
<<
"WARNING: The double buffer channel has been closed. The "
"prefetch thread will terminate."
;
break
;
...
...
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