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
兽拳
Paddle
提交
399d3a2d
P
Paddle
项目概览
兽拳
/
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 >>
未验证
提交
399d3a2d
编写于
11月 29, 2017
作者:
Q
qingqing01
提交者:
GitHub
11月 29, 2017
浏览文件
操作
浏览文件
下载
差异文件
Merge pull request #6022 from qingqing01/nvprof
Small fix in profiler.py of fluid.
上级
0066bbba
a1bf4c28
变更
2
隐藏空白更改
内联
并排
Showing
2 changed file
with
6 addition
and
6 deletion
+6
-6
python/paddle/v2/fluid/profiler.py
python/paddle/v2/fluid/profiler.py
+5
-5
python/paddle/v2/fluid/tests/test_profiler.py
python/paddle/v2/fluid/tests/test_profiler.py
+1
-1
未找到文件。
python/paddle/v2/fluid/profiler.py
浏览文件 @
399d3a2d
...
@@ -22,21 +22,21 @@ def cuda_profiler(output_file, output_mode=None, config=None):
...
@@ -22,21 +22,21 @@ def cuda_profiler(output_file, output_mode=None, config=None):
`output_file` with Key-Value pair format or Comma separated values format.
`output_file` with Key-Value pair format or Comma separated values format.
The user can set the output mode by `output_mode` argument and set the
The user can set the output mode by `output_mode` argument and set the
counters/options for profiling by `config` argument. The default config
counters/options for profiling by `config` argument. The default config
caontains
'gpustarttimestamp', 'gpustarttimestamp', 'gridsize3d',
is [
'gpustarttimestamp', 'gpustarttimestamp', 'gridsize3d',
'threadblocksize', 'streamid', 'enableonstart 0', 'conckerneltrace'.
'threadblocksize', 'streamid', 'enableonstart 0', 'conckerneltrace'
]
.
Args:
Args:
output_file (string) : The output file name, the result will be
output_file (string) : The output file name, the result will be
written into this file.
written into this file.
output_mode (string) : The output mode has Key-Value pair format and
output_mode (string) : The output mode has Key-Value pair format and
Comma separated values format. It should be 'kv' or 'csv'.
Comma separated values format. It should be 'kv
p
' or 'csv'.
config (string) : The profiler options and counters can refer to
config (string) : The profiler options and counters can refer to
"Compute Command Line Profiler User Guide".
"Compute Command Line Profiler User Guide".
"""
"""
if
output_mode
is
None
:
if
output_mode
is
None
:
output_mode
=
'csv'
output_mode
=
'csv'
if
output_mode
not
in
[
'kv'
,
'csv'
]:
if
output_mode
not
in
[
'kv
p
'
,
'csv'
]:
raise
ValueError
(
"The output mode must be 'k
ey-value
' or 'csv'."
)
raise
ValueError
(
"The output mode must be 'k
vp
' or 'csv'."
)
config
=
NVPROF_CONFIG
if
config
is
None
else
config
config
=
NVPROF_CONFIG
if
config
is
None
else
config
core
.
nvprof_init
(
output_file
,
output_mode
,
config
)
core
.
nvprof_init
(
output_file
,
output_mode
,
config
)
# Enables profiler collection by the active CUDA profiling tool.
# Enables profiler collection by the active CUDA profiling tool.
...
...
python/paddle/v2/fluid/tests/test_profiler.py
浏览文件 @
399d3a2d
...
@@ -20,7 +20,7 @@ class TestProfiler(unittest.TestCase):
...
@@ -20,7 +20,7 @@ class TestProfiler(unittest.TestCase):
with
profiler
.
cuda_profiler
(
"cuda_profiler.txt"
,
'csv'
)
as
nvprof
:
with
profiler
.
cuda_profiler
(
"cuda_profiler.txt"
,
'csv'
)
as
nvprof
:
for
i
in
range
(
epoc
):
for
i
in
range
(
epoc
):
input
=
np
.
random
.
random
(
dshape
).
astype
(
"float32"
)
input
=
np
.
random
.
random
(
dshape
).
astype
(
'float32'
)
exe
.
run
(
fluid
.
default_main_program
(),
feed
=
{
'data'
:
input
})
exe
.
run
(
fluid
.
default_main_program
(),
feed
=
{
'data'
:
input
})
...
...
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