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
afeixing77
Paddle
提交
fb6697b4
P
Paddle
项目概览
afeixing77
/
Paddle
与 Fork 源项目一致
Fork自
PaddlePaddle / Paddle
通知
1
Star
0
Fork
0
代码
文件
提交
分支
Tags
贡献者
分支图
Diff
Issue
0
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
0
Issue
0
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
fb6697b4
编写于
12月 15, 2020
作者:
S
ShenLiang
提交者:
GitHub
12月 15, 2020
浏览文件
操作
浏览文件
下载
电子邮件补丁
差异文件
Fix the dowanload bug in the case of multiple machines (#29551)
* fix the dowanload bug * add sort for ips
上级
1efef8ba
变更
1
隐藏空白更改
内联
并排
Showing
1 changed file
with
21 addition
and
3 deletion
+21
-3
python/paddle/utils/download.py
python/paddle/utils/download.py
+21
-3
未找到文件。
python/paddle/utils/download.py
浏览文件 @
fb6697b4
...
...
@@ -140,6 +140,21 @@ def _map_path(url, root_dir):
return
osp
.
join
(
root_dir
,
fpath
)
def
_get_unique_endpoints
(
trainer_endpoints
):
# Sorting is to avoid different environmental variables for each card
trainer_endpoints
.
sort
()
ips
=
set
()
unique_endpoints
=
set
()
for
endpoint
in
trainer_endpoints
:
ip
=
endpoint
.
split
(
":"
)[
0
]
if
ip
in
ips
:
continue
ips
.
add
(
ip
)
unique_endpoints
.
add
(
endpoint
)
logger
.
info
(
"unique_endpoints {}"
.
format
(
unique_endpoints
))
return
unique_endpoints
def
get_path_from_url
(
url
,
root_dir
,
md5sum
=
None
,
check_exist
=
True
):
""" Download from given url to root_dir.
if file or directory specified by url is exists under
...
...
@@ -161,17 +176,20 @@ def get_path_from_url(url, root_dir, md5sum=None, check_exist=True):
assert
is_url
(
url
),
"downloading from {} not a url"
.
format
(
url
)
# parse path after download to decompress under root_dir
fullpath
=
_map_path
(
url
,
root_dir
)
# Mainly used to solve the problem of downloading data from different
# machines in the case of multiple machines. Different ips will download
# data, and the same ip will only download data once.
unique_endpoints
=
_get_unique_endpoints
(
ParallelEnv
().
trainer_endpoints
[:])
if
osp
.
exists
(
fullpath
)
and
check_exist
and
_md5check
(
fullpath
,
md5sum
):
logger
.
info
(
"Found {}"
.
format
(
fullpath
))
else
:
if
ParallelEnv
().
local_rank
==
0
:
if
ParallelEnv
().
current_endpoint
in
unique_endpoints
:
fullpath
=
_download
(
url
,
root_dir
,
md5sum
)
else
:
while
not
os
.
path
.
exists
(
fullpath
):
time
.
sleep
(
1
)
if
ParallelEnv
().
local_rank
==
0
:
if
ParallelEnv
().
current_endpoint
in
unique_endpoints
:
if
tarfile
.
is_tarfile
(
fullpath
)
or
zipfile
.
is_zipfile
(
fullpath
):
fullpath
=
_decompress
(
fullpath
)
...
...
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