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
BaiXuePrincess
Paddle
提交
826c607e
P
Paddle
项目概览
BaiXuePrincess
/
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
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
未验证
提交
826c607e
编写于
7月 02, 2020
作者:
K
Kaipeng Deng
提交者:
GitHub
7月 02, 2020
浏览文件
操作
浏览文件
下载
电子邮件补丁
差异文件
Fix test multiprocess dataloader static (#25287)
* fix test_multiprocess_dataloader_static random fail. test=develop
上级
4474fc10
变更
3
隐藏空白更改
内联
并排
Showing
3 changed file
with
15 addition
and
16 deletion
+15
-16
python/paddle/fluid/dataloader/dataloader_iter.py
python/paddle/fluid/dataloader/dataloader_iter.py
+10
-2
python/paddle/fluid/tests/unittests/test_multiprocess_dataloader_dynamic.py
...d/tests/unittests/test_multiprocess_dataloader_dynamic.py
+1
-6
python/paddle/fluid/tests/unittests/test_multiprocess_dataloader_static.py
...id/tests/unittests/test_multiprocess_dataloader_static.py
+4
-8
未找到文件。
python/paddle/fluid/dataloader/dataloader_iter.py
浏览文件 @
826c607e
...
...
@@ -128,8 +128,10 @@ class _DataLoaderIterSingleProcess(_DataLoaderIterBase):
self
.
_need_check_feed
=
[
v
.
desc
.
need_check_feed
()
for
v
in
self
.
_feed_list
]
# if only 1 place, do not need to keep order
self
.
_blocking_queue
=
core
.
init_lod_tensor_blocking_queue
(
core
.
Variable
(),
self
.
_blocking_queue_capacity
,
True
)
core
.
Variable
(),
self
.
_blocking_queue_capacity
,
len
(
self
.
_places
)
>
1
)
self
.
_reader
=
core
.
create_py_reader
(
self
.
_blocking_queue
,
self
.
_var_names
,
self
.
_shapes
,
self
.
_dtypes
,
self
.
_need_check_feed
,
self
.
_places
,
self
.
_use_buffer_reader
,
True
)
...
...
@@ -280,8 +282,9 @@ class _DataLoaderIterMultiProcess(_DataLoaderIterBase):
self
.
_need_check_feed
=
[
v
.
desc
.
need_check_feed
()
for
v
in
self
.
_feed_list
]
# if only 1 place, do not need to keep order
self
.
_blocking_queue
=
core
.
init_lod_tensor_blocking_queue
(
core
.
Variable
(),
self
.
_outstanding_capacity
,
True
)
core
.
Variable
(),
self
.
_outstanding_capacity
,
len
(
self
.
_places
)
>
1
)
self
.
_reader
=
core
.
create_py_reader
(
self
.
_blocking_queue
,
self
.
_var_names
,
self
.
_shapes
,
self
.
_dtypes
,
self
.
_need_check_feed
,
self
.
_places
,
self
.
_use_buffer_reader
,
True
)
...
...
@@ -442,6 +445,11 @@ class _DataLoaderIterMultiProcess(_DataLoaderIterBase):
# get data again
data
=
self
.
_data_queue
.
get
(
timeout
=
self
.
_timeout
)
except
Exception
as
e
:
# check if thread done event set when waiting data
if
self
.
_thread_done_event
.
is_set
():
continue
# check failed workers
failed_workers
=
[]
for
i
,
w
in
enumerate
(
self
.
_workers
):
if
self
.
_worker_status
[
i
]
and
not
w
.
is_alive
():
...
...
python/paddle/fluid/tests/unittests/test_multiprocess_dataloader_dynamic.py
浏览文件 @
826c607e
...
...
@@ -28,12 +28,7 @@ from paddle.fluid.dygraph.nn import Linear
from
paddle.fluid.dygraph.base
import
to_variable
from
test_multiprocess_dataloader_static
import
RandomDataset
,
prepare_places
EPOCH_NUM
=
5
BATCH_SIZE
=
16
IMAGE_SIZE
=
784
SAMPLE_NUM
=
400
CLASS_NUM
=
10
from
test_multiprocess_dataloader_static
import
EPOCH_NUM
,
BATCH_SIZE
,
IMAGE_SIZE
,
SAMPLE_NUM
,
CLASS_NUM
class
SimpleFCNet
(
fluid
.
dygraph
.
Layer
):
...
...
python/paddle/fluid/tests/unittests/test_multiprocess_dataloader_static.py
浏览文件 @
826c607e
...
...
@@ -25,10 +25,10 @@ import numpy as np
import
paddle.fluid
as
fluid
from
paddle.io
import
Dataset
,
BatchSampler
,
DataLoader
EPOCH_NUM
=
5
BATCH_SIZE
=
16
IMAGE_SIZE
=
784
SAMPLE_NUM
=
4
00
EPOCH_NUM
=
3
BATCH_SIZE
=
8
IMAGE_SIZE
=
32
SAMPLE_NUM
=
1
00
CLASS_NUM
=
10
...
...
@@ -157,10 +157,6 @@ class TestStaticDataLoader(unittest.TestCase):
return
ret
def
test_main
(
self
):
# FIXME(dkp): disable for random fail in Py35 cloud,
# should be fixed ASAP
if
sys
.
version
[:
3
]
==
'3.5'
:
return
for
p
in
prepare_places
(
True
):
results
=
[]
for
num_workers
in
[
0
,
2
]:
...
...
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