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
PaddlePaddle
PaddleDetection
提交
2240f7fd
P
PaddleDetection
项目概览
PaddlePaddle
/
PaddleDetection
大约 1 年 前同步成功
通知
695
Star
11112
Fork
2696
代码
文件
提交
分支
Tags
贡献者
分支图
Diff
Issue
184
列表
看板
标记
里程碑
合并请求
40
Wiki
0
Wiki
分析
仓库
DevOps
项目成员
Pages
P
PaddleDetection
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
184
Issue
184
列表
看板
标记
里程碑
合并请求
40
合并请求
40
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
2240f7fd
编写于
6月 28, 2019
作者:
Z
Zeng Jinle
提交者:
GitHub
6月 28, 2019
浏览文件
操作
浏览文件
下载
电子邮件补丁
差异文件
refine error msg of CUDAPlace, test=release/1.5 (#18376)
上级
924e53b7
变更
1
隐藏空白更改
内联
并排
Showing
1 changed file
with
34 addition
and
5 deletion
+34
-5
paddle/fluid/pybind/pybind.cc
paddle/fluid/pybind/pybind.cc
+34
-5
未找到文件。
paddle/fluid/pybind/pybind.cc
浏览文件 @
2240f7fd
...
...
@@ -13,6 +13,7 @@ See the License for the specific language governing permissions and
limitations under the License. */
#include <Python.h>
#include <algorithm>
#include <cstdlib>
#include <map>
#include <memory>
#include <mutex> // NOLINT // for call_once
...
...
@@ -786,13 +787,41 @@ All parameter, weight, gradient are variables in Paddle.
.
def
(
"__init__"
,
[](
platform
::
CUDAPlace
&
self
,
int
dev_id
)
{
#ifdef PADDLE_WITH_CUDA
PADDLE_ENFORCE
(
dev_id
>=
0
&&
dev_id
<
platform
::
GetCUDADeviceCount
(),
"Invalid CUDAPlace(%d), must inside [0, %d)"
,
dev_id
,
platform
::
GetCUDADeviceCount
());
if
(
UNLIKELY
(
dev_id
<
0
))
{
LOG
(
ERROR
)
<<
string
::
Sprintf
(
"Invalid CUDAPlace(%d), device id must be 0 or "
"positive integer"
,
dev_id
);
std
::
exit
(
-
1
);
}
if
(
UNLIKELY
(
dev_id
>=
platform
::
GetCUDADeviceCount
()))
{
if
(
platform
::
GetCUDADeviceCount
()
==
0
)
{
LOG
(
ERROR
)
<<
"Cannot use GPU because there is no GPU "
"detected on your "
"machine."
;
std
::
exit
(
-
1
);
}
else
{
LOG
(
ERROR
)
<<
string
::
Sprintf
(
"Invalid CUDAPlace(%d), must inside [0, %d), because GPU "
"number on your machine is %d"
,
dev_id
,
platform
::
GetCUDADeviceCount
(),
platform
::
GetCUDADeviceCount
());
std
::
exit
(
-
1
);
}
}
new
(
&
self
)
platform
::
CUDAPlace
(
dev_id
);
#else
PADDLE_THROW
(
"Cannot use CUDAPlace in CPU only version"
);
LOG
(
ERROR
)
<<
string
::
Sprintf
(
"Cannot use GPU because you have installed CPU version "
"PaddlePaddle.
\n
"
"If you want to use GPU, please try to install GPU version "
"PaddlePaddle by: pip install paddlepaddle-gpu
\n
"
"If you only have CPU, please change CUDAPlace(%d) to be "
"CPUPlace().
\n
"
,
dev_id
);
std
::
exit
(
-
1
);
#endif
})
.
def
(
"_type"
,
&
PlaceIndex
<
platform
::
CUDAPlace
>
)
...
...
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