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
Paddle
提交
3587d921
P
Paddle
项目概览
PaddlePaddle
/
Paddle
大约 1 年 前同步成功
通知
2299
Star
20931
Fork
5422
代码
文件
提交
分支
Tags
贡献者
分支图
Diff
Issue
1423
列表
看板
标记
里程碑
合并请求
543
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
1,423
Issue
1,423
列表
看板
标记
里程碑
合并请求
543
合并请求
543
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
3587d921
编写于
6月 19, 2018
作者:
G
gongweibao
提交者:
GitHub
6月 19, 2018
浏览文件
操作
浏览文件
下载
电子邮件补丁
差异文件
Fix kube jobs generator's bugs. (#11557)
上级
1d7e60fd
变更
3
隐藏空白更改
内联
并排
Showing
3 changed file
with
25 addition
and
7 deletion
+25
-7
benchmark/fluid/Dockerfile
benchmark/fluid/Dockerfile
+13
-4
benchmark/fluid/fluid_benchmark.py
benchmark/fluid/fluid_benchmark.py
+9
-0
benchmark/fluid/kube_gen_job.py
benchmark/fluid/kube_gen_job.py
+3
-3
未找到文件。
benchmark/fluid/Dockerfile
浏览文件 @
3587d921
FROM
nvidia/cuda:9.0-cudnn7-devel-ubuntu16.04
# Use UBUNTU_MIRROR can speed up apt-get speed.
# ARG UBUNTU_MIRROR
# RUN /bin/bash -c 'if [[ -n ${UBUNTU_MIRROR} ]]; then sed -i 's#http://archive.ubuntu.com/ubuntu#${UBUNTU_MIRROR}#g' /etc/apt/sources.list; fi'
RUN
apt-get update
&&
apt-get
install
-y
python python-pip iputils-ping libgtk2.0-dev wget vim net-tools iftop python-opencv
RUN
ln
-s
/usr/lib/x86_64-linux-gnu/libcudnn.so.7 /usr/lib/libcudnn.so
&&
ln
-s
/usr/lib/x86_64-linux-gnu/libnccl.so.2 /usr/lib/libnccl.so
RUN
pip
install
-U
pip
RUN
pip
install
-U
kubernetes paddlepaddle
# IMPORTANT:
# Add "ENV http_proxy=http://ip:port" if your download is slow, and don't forget to unset it at runtime.
# exmaple: unset http_proxy && unset https_proxy && python fluid_benchmark.py ...
RUN
pip
install
-U
pip
RUN
pip
install
-U
kubernetes paddlepaddle
RUN
sh
-c
'echo "import paddle.v2 as paddle\npaddle.dataset.cifar.train10()\npaddle.dataset.flowers.fetch()" | python'
RUN
sh
-c
'echo "import paddle.v2 as paddle\npaddle.dataset.mnist.train()\npaddle.dataset.mnist.test()\npaddle.dataset.imdb.fetch()" | python'
...
...
@@ -14,9 +21,11 @@ RUN pip uninstall -y paddlepaddle && mkdir /workspace
ADD
https://raw.githubusercontent.com/PaddlePaddle/cloud/develop/docker/paddle_k8s /usr/bin
ADD
https://raw.githubusercontent.com/PaddlePaddle/cloud/develop/docker/k8s_tools.py /root
RUN
chmod
+x /usr/bin/paddle_k8s
ADD
*.whl /
RUN
pip
install
/
*
.whl
&&
rm
-f
/
*
.whl
&&
chmod
+x /usr/bin/paddle_k8s
RUN
pip
install
/
*
.whl
&&
rm
-f
/
*
.whl
ENV
LD_LIBRARY_PATH=/usr/local/lib
ADD
fluid_benchmark.py recordio_converter.py models/ /workspace/
ADD
fluid_benchmark.py recordio_converter.py args.py recordio_converter.py run.sh run_fluid_benchmark.sh /workspace/
ADD
models/ /workspace/models/
benchmark/fluid/fluid_benchmark.py
浏览文件 @
3587d921
...
...
@@ -301,9 +301,18 @@ def print_train_time(start_time, end_time, num_samples):
(
num_samples
,
train_elapsed
,
examples_per_sec
))
def
print_paddle_envs
():
print
(
'----------- Configuration envs -----------'
)
for
k
in
os
.
environ
:
if
"PADDLE_"
in
k
:
print
"ENV %s:%s"
%
(
k
,
os
.
environ
[
k
])
print
(
'------------------------------------------------'
)
def
main
():
args
=
parse_args
()
print_arguments
(
args
)
print_paddle_envs
()
# the unique trainer id, starting from 0, needed by trainer
# only
...
...
benchmark/fluid/kube_gen_job.py
浏览文件 @
3587d921
...
...
@@ -17,6 +17,7 @@ import copy
import
argparse
import
random
import
os
import
copy
from
kube_templates
import
pserver
,
trainer
,
envs
...
...
@@ -109,10 +110,9 @@ def gen_job():
envs
.
append
({
"name"
:
"PADDLE_JOB_NAME"
,
"value"
:
args
.
jobname
})
envs
.
append
({
"name"
:
"PADDLE_TRAINERS"
,
"value"
:
str
(
args
.
trainers
)})
envs
.
append
({
"name"
:
"PSERVERS"
,
"value"
:
str
(
args
.
pservers
)})
envs
.
append
({
"name"
:
"P
ADDLE_P
SERVERS"
,
"value"
:
str
(
args
.
pservers
)})
envs
.
append
({
"name"
:
"ENTRY"
,
"value"
:
args
.
entry
})
envs
.
append
({
"name"
:
"PADDLE_PSERVER_PORT"
,
"value"
:
str
(
args
.
port
)})
envs
.
append
({
"name"
:
"PADDLE_PSERVER_PORT"
,
"value"
:
str
(
args
.
port
)})
# NOTE: these directories below are cluster specific, please modify
# this settings before you run on your own cluster.
envs
.
append
({
...
...
@@ -166,7 +166,7 @@ def gen_job():
tn
[
"spec"
][
"template"
][
"spec"
][
"volumes"
]
=
volumes
tn_container
[
"volumeMounts"
]
=
volumeMounts
ps_container
[
"env"
]
=
envs
ps_container
[
"env"
]
=
copy
.
deepcopy
(
envs
)
ps_container
[
"env"
].
append
({
"name"
:
"PADDLE_TRAINING_ROLE"
,
"value"
:
"PSERVER"
...
...
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