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
438ca7f6
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
438ca7f6
编写于
7月 21, 2022
作者:
Z
zhaoyingli
提交者:
GitHub
7月 21, 2022
浏览文件
操作
浏览文件
下载
电子邮件补丁
差异文件
[AutoParallel] fix unittest with paddle.distributed.launch (#44439)
* fix unittest * fix log_dir * _enable_legacy_dygraph
上级
98e96853
变更
8
隐藏空白更改
内联
并排
Showing
8 changed file
with
21 addition
and
18 deletion
+21
-18
python/paddle/distributed/auto_parallel/process_group.py
python/paddle/distributed/auto_parallel/process_group.py
+4
-1
python/paddle/fluid/tests/unittests/auto_parallel/test_auto_parallel_relaunch.py
...ts/unittests/auto_parallel/test_auto_parallel_relaunch.py
+1
-1
python/paddle/fluid/tests/unittests/auto_parallel/test_converter.py
...dle/fluid/tests/unittests/auto_parallel/test_converter.py
+5
-5
python/paddle/fluid/tests/unittests/auto_parallel/test_engine_api.py
...le/fluid/tests/unittests/auto_parallel/test_engine_api.py
+2
-2
python/paddle/fluid/tests/unittests/auto_parallel/test_engine_api_dp.py
...fluid/tests/unittests/auto_parallel/test_engine_api_dp.py
+2
-2
python/paddle/fluid/tests/unittests/auto_parallel/test_high_order_grad.py
...uid/tests/unittests/auto_parallel/test_high_order_grad.py
+5
-5
python/paddle/fluid/tests/unittests/auto_parallel/test_relaunch_with_gpt_planner.py
...unittests/auto_parallel/test_relaunch_with_gpt_planner.py
+1
-1
python/paddle/fluid/tests/unittests/auto_parallel/test_relaunch_with_planner.py
...sts/unittests/auto_parallel/test_relaunch_with_planner.py
+1
-1
未找到文件。
python/paddle/distributed/auto_parallel/process_group.py
浏览文件 @
438ca7f6
...
...
@@ -16,10 +16,12 @@ from collections import OrderedDict
import
paddle
import
paddle.fluid.core
as
core
from
..collective
import
_get_global_env
from
..collective
import
_new_ring_id
from
...fluid.framework
import
_non_static_mode
from
...fluid.layers.tensor
import
fill_constant
from
paddle.fluid.framework
import
_enable_legacy_dygraph
def
get_all_process_groups
():
...
...
@@ -134,7 +136,8 @@ class ProcessGroup:
# TODO(shenliang03): This is a temporary solution to solve the problem of
# hang caused by cross-creation of new_group
paddle
.
framework
.
_in_legacy_dygraph
()
paddle
.
disable_static
()
_enable_legacy_dygraph
()
paddle
.
set_device
(
'gpu:%d'
%
paddle
.
distributed
.
ParallelEnv
().
dev_id
)
tmp
=
paddle
.
to_tensor
(
...
...
python/paddle/fluid/tests/unittests/auto_parallel/test_auto_parallel_relaunch.py
浏览文件 @
438ca7f6
...
...
@@ -126,7 +126,7 @@ class TestAutoParallelReLaunch(unittest.TestCase):
coverage_args
=
[]
cmd
=
[
sys
.
executable
,
"-u"
]
+
coverage_args
+
[
"-m"
,
"launch"
,
"--log_dir"
,
self
.
temp_dir
.
name
,
"-m"
,
"
paddle.distributed.
launch"
,
"--log_dir"
,
self
.
temp_dir
.
name
,
"--cluster_topo_path"
,
cluster_json_path
,
"--rank_mapping_path"
,
mapping_json_path
,
"--enable_auto_mapping"
,
"True"
,
launch_model_path
...
...
python/paddle/fluid/tests/unittests/auto_parallel/test_converter.py
浏览文件 @
438ca7f6
...
...
@@ -12,6 +12,7 @@
# See the License for the specific language governing permissions and
# limitations under the License.
import
tempfile
import
unittest
import
os
import
sys
...
...
@@ -32,18 +33,17 @@ class TestConverter(unittest.TestCase):
else
:
coverage_args
=
[]
tmp_dir
=
tempfile
.
TemporaryDirectory
()
cmd
=
[
sys
.
executable
,
"-u"
]
+
coverage_args
+
[
"-m"
,
"launch"
,
"--gpus"
,
"0,1"
,
launch_model_path
"-m"
,
"paddle.distributed.launch"
,
"--devices"
,
"0,1"
,
"--log_dir"
,
tmp_dir
.
name
,
launch_model_path
]
process
=
subprocess
.
Popen
(
cmd
)
process
.
wait
()
self
.
assertEqual
(
process
.
returncode
,
0
)
# Remove unnecessary files
log_path
=
os
.
path
.
join
(
file_dir
,
"log"
)
if
os
.
path
.
exists
(
log_path
):
shutil
.
rmtree
(
log_path
)
tmp_dir
.
cleanup
()
def
test_input_invalid
(
self
):
with
self
.
assertRaises
(
ValueError
):
...
...
python/paddle/fluid/tests/unittests/auto_parallel/test_engine_api.py
浏览文件 @
438ca7f6
...
...
@@ -34,8 +34,8 @@ class TestEngineAPI(unittest.TestCase):
tmp_dir
=
tempfile
.
TemporaryDirectory
()
cmd
=
[
sys
.
executable
,
"-u"
]
+
coverage_args
+
[
"-m"
,
"
launch"
,
"--gpus"
,
"0,1"
,
"--log_dir"
,
tmp_dir
.
name
,
launch_model_path
"-m"
,
"
paddle.distributed.launch"
,
"--devices"
,
"0,1"
,
"--log_dir"
,
tmp_dir
.
name
,
launch_model_path
]
process
=
subprocess
.
Popen
(
cmd
)
...
...
python/paddle/fluid/tests/unittests/auto_parallel/test_engine_api_dp.py
浏览文件 @
438ca7f6
...
...
@@ -34,8 +34,8 @@ class TestEngineAPI(unittest.TestCase):
tmp_dir
=
tempfile
.
TemporaryDirectory
()
cmd
=
[
sys
.
executable
,
"-u"
]
+
coverage_args
+
[
"-m"
,
"
launch"
,
"--gpus"
,
"0,1"
,
"--log_dir"
,
tmp_dir
.
name
,
launch_model_path
"-m"
,
"
paddle.distributed.launch"
,
"--devices"
,
"0,1"
,
"--log_dir"
,
tmp_dir
.
name
,
launch_model_path
]
process
=
subprocess
.
Popen
(
cmd
)
...
...
python/paddle/fluid/tests/unittests/auto_parallel/test_high_order_grad.py
浏览文件 @
438ca7f6
...
...
@@ -12,6 +12,7 @@
# See the License for the specific language governing permissions and
# limitations under the License.
import
tempfile
import
unittest
import
os
import
sys
...
...
@@ -31,18 +32,17 @@ class TestHighOrderGrad(unittest.TestCase):
else
:
coverage_args
=
[]
tmp_dir
=
tempfile
.
TemporaryDirectory
()
cmd
=
[
sys
.
executable
,
"-u"
]
+
coverage_args
+
[
"-m"
,
"launch"
,
"--gpus"
,
"0,1"
,
launch_model_path
"-m"
,
"paddle.distributed.launch"
,
"--devices"
,
"0,1"
,
"--log_dir"
,
tmp_dir
.
name
,
launch_model_path
]
process
=
subprocess
.
Popen
(
cmd
)
process
.
wait
()
self
.
assertEqual
(
process
.
returncode
,
0
)
# Remove unnecessary files
log_path
=
os
.
path
.
join
(
file_dir
,
"log"
)
if
os
.
path
.
exists
(
log_path
):
shutil
.
rmtree
(
log_path
)
tmp_dir
.
cleanup
()
if
__name__
==
"__main__"
:
...
...
python/paddle/fluid/tests/unittests/auto_parallel/test_relaunch_with_gpt_planner.py
浏览文件 @
438ca7f6
...
...
@@ -56,7 +56,7 @@ class TestPlannerReLaunch(unittest.TestCase):
coverage_args
=
[]
cmd
=
[
sys
.
executable
,
"-u"
]
+
coverage_args
+
[
"-m"
,
"launch"
,
"--log_dir"
,
self
.
temp_dir
.
name
,
"-m"
,
"
paddle.distributed.
launch"
,
"--log_dir"
,
self
.
temp_dir
.
name
,
"--cluster_topo_path"
,
cluster_json_path
,
"--rank_mapping_path"
,
mapping_json_path
,
"--enable_auto_mapping"
,
"True"
,
launch_model_path
...
...
python/paddle/fluid/tests/unittests/auto_parallel/test_relaunch_with_planner.py
浏览文件 @
438ca7f6
...
...
@@ -56,7 +56,7 @@ class TestPlannerReLaunch(unittest.TestCase):
coverage_args
=
[]
cmd
=
[
sys
.
executable
,
"-u"
]
+
coverage_args
+
[
"-m"
,
"launch"
,
"--log_dir"
,
self
.
temp_dir
.
name
,
"-m"
,
"
paddle.distributed.
launch"
,
"--log_dir"
,
self
.
temp_dir
.
name
,
"--cluster_topo_path"
,
cluster_json_path
,
"--rank_mapping_path"
,
mapping_json_path
,
"--enable_auto_mapping"
,
"True"
,
launch_model_path
...
...
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