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
03a7ec55
P
Paddle
项目概览
PaddlePaddle
/
Paddle
大约 1 年 前同步成功
通知
2298
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
体验新版 GitCode,发现更多精彩内容 >>
未验证
提交
03a7ec55
编写于
4月 16, 2019
作者:
G
guru4elephant
提交者:
GitHub
4月 16, 2019
浏览文件
操作
浏览文件
下载
差异文件
Merge pull request #16746 from xjqbest/dataset_merge_develop
move split filelist from trainer.py to fleet & fix error
上级
5d45eb06
7968887f
变更
3
隐藏空白更改
内联
并排
Showing
3 changed file
with
57 addition
and
1 deletion
+57
-1
python/paddle/fluid/dataset.py
python/paddle/fluid/dataset.py
+21
-0
python/paddle/fluid/incubate/fleet/base/role_maker.py
python/paddle/fluid/incubate/fleet/base/role_maker.py
+1
-1
python/paddle/fluid/incubate/fleet/parameter_server/__init__.py
.../paddle/fluid/incubate/fleet/parameter_server/__init__.py
+35
-0
未找到文件。
python/paddle/fluid/dataset.py
浏览文件 @
03a7ec55
...
...
@@ -219,6 +219,7 @@ class InMemoryDataset(DatasetBase):
>>> dataset = fluid.DatasetFactory.create_dataset("InMemoryDataset")
>>> filelist = ["a.txt", "b.txt"]
>>> dataset.set_filelist(filelist)
>>> dataset.load_into_memory()
>>> dataset.local_shuffle()
"""
self
.
dataset
.
local_shuffle
()
...
...
@@ -236,6 +237,7 @@ class InMemoryDataset(DatasetBase):
>>> dataset = fluid.DatasetFactory.create_dataset("InMemoryDataset")
>>> filelist = ["a.txt", "b.txt"]
>>> dataset.set_filelist(filelist)
>>> dataset.load_into_memory()
>>> dataset.global_shuffle(fleet)
Args:
...
...
@@ -255,6 +257,25 @@ class InMemoryDataset(DatasetBase):
if
fleet
is
not
None
:
fleet
.
fleet_instance
.
role_maker_
.
_barrier_worker
()
def
release_memory
(
self
):
"""
Release InMemoryDataset memory data, when data will not be used again.
Example:
>>> import paddle.fluid as fluid
>>> import paddle.fluid.incubate.fleet.parameter_server as fleet
>>> dataset = fluid.DatasetFactory.create_dataset("InMemoryDataset")
>>> filelist = ["a.txt", "b.txt"]
>>> dataset.set_filelist(filelist)
>>> dataset.load_into_memory()
>>> dataset.global_shuffle(fleet)
>>> exe = fluid.Executor(fluid.CPUPlace())
>>> exe.run(fluid.default_startup_program())
>>> exe.train_from_dataset(fluid.default_main_program(), dataset)
>>> dataset.release_memory()
"""
self
.
dataset
.
release_memory
()
class
QueueDataset
(
DatasetBase
):
"""
...
...
python/paddle/fluid/incubate/fleet/base/role_maker.py
浏览文件 @
03a7ec55
...
...
@@ -128,7 +128,7 @@ class MPIRoleMaker(RoleMakerBase):
"""
finalize the current MPI instance.
"""
self
.
_comm
.
finalize
()
pass
class
MPISymetricRoleMaker
(
MPIRoleMaker
):
...
...
python/paddle/fluid/incubate/fleet/parameter_server/__init__.py
浏览文件 @
03a7ec55
...
...
@@ -241,6 +241,40 @@ class Fleet(object):
"""
self
.
_fleet_ptr
.
save_model
(
save_path
)
def
split_filelist
(
self
,
filelist
):
"""
split filelist before distributed training,
for example, filelist is [a, b, c ,d, e] and trainer_num = 2,
then trainer 0 gets [a, b, c] and trainer 1 gets [d, e]
Example:
>>> all_filelist = ["a.txt", "b.txt", "c.txt"]
>>> my_filelist = fleet.split_filelist(all_filelist)
>>> dataset = fluid.DatasetFactory().create_dataset()
>>> dataset.set_filelist(my_filelist)
Args:
filelist(list): list of filename, can be local or hdfs/afs.
Returns:
list of filename which belongs to this trainer.
"""
file_num
=
len
(
filelist
)
trainer_id
=
self
.
get_worker_index
()
trainer_num
=
self
.
get_worker_num
()
if
trainer_num
>
file_num
:
raise
ValueError
(
"trainer_num should be <= file_num : "
"%s > %s"
%
(
trainer_num
,
file_num
))
# get interval of filelist, it's [ )
start
=
0
end
=
0
for
i
in
range
(
0
,
trainer_id
+
1
):
length
=
file_num
/
trainer_num
+
(
i
<
(
file_num
%
trainer_num
))
start
=
end
end
+=
length
my_filelist
=
filelist
[
start
:
end
]
return
my_filelist
def
_set_opt_info
(
self
,
opt_info
):
"""
this function saves the result from DistributedOptimizer.minimize()
...
...
@@ -337,3 +371,4 @@ save_pserver_model = fleet_instance.save_pserver_model
worker_num
=
fleet_instance
.
get_worker_num
server_num
=
fleet_instance
.
get_server_num
worker_index
=
fleet_instance
.
get_worker_index
split_filelist
=
fleet_instance
.
split_filelist
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