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
DMLL
Paddle
提交
8b70f0f3
P
Paddle
项目概览
DMLL
/
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 >>
提交
8b70f0f3
编写于
2月 13, 2017
作者:
Y
Yi Wang
浏览文件
操作
浏览文件
下载
电子邮件补丁
差异文件
addrss
https://github.com/PaddlePaddle/Paddle/issues/1319
as well
上级
5a1f0617
变更
1
显示空白变更内容
内联
并排
Showing
1 changed file
with
12 addition
and
6 deletion
+12
-6
doc/design/api.md
doc/design/api.md
+12
-6
未找到文件。
doc/design/api.md
浏览文件 @
8b70f0f3
...
...
@@ -166,14 +166,19 @@ There are some open questions here:
### Training
The recommended way to training a model is to call
`paddle.train`
,
which simply calls
`paddle.
optimiz
er.Default`
, a global variable of
type
`paddle.
optimiz
er.SGD`
. Equivalently, we can do
which simply calls
`paddle.
train
er.Default`
, a global variable of
type
`paddle.
train
er.SGD`
. Equivalently, we can do
```
python
opt
=
paddle
.
optimizer
.
SGD
(...
)
opt
=
paddle
.
trainer
.
SGD
(...,
paddle
.
updater
.
Adam
(...)
)
opt
.
train
(
model
,
reader
=
read
,
...)
```
Please be aware that a trainer requires an updater as its data
member. This is to make it easier to customize trainers, as
discussed
[
here
](
https://github.com/PaddlePaddle/Paddle/issues/1319
)
.
#### Distributed Training
If users want to do distributed training on a cluster, s/he should
...
...
@@ -185,8 +190,9 @@ access a Kubernetes cluster, s/he should be able to call
```
python
paddle
.
dist_train
(
model
,
trainer
=
paddle
.
trainer
.
SGD
(...,
paddle
.
updater
.
Adam
(...)),
reader
=
read
,
optimizer
=
paddle
.
optimizer
.
SGDOptimizer
(...),
k8s_user
=
"yi"
,
k8s_token
=
"kube_cluster_tls.pem"
,
k8s_job
=
"hello"
,
...
...
@@ -196,7 +202,7 @@ paddle.dist_train(model,
The pseudo code if
`paddle.dist_train`
is as follows:
```
python
def
dist_train
():
def
dist_train
(
topology
,
parameters
,
trainer
,
reader
,
...
):
if
os
.
getenv
(
"KUBERNETES_SERVICE_HOST"
)
==
None
:
image_name
=
k8s_user
+
'/'
+
k8s_job
docker_build
(
image_name
)
...
...
@@ -209,7 +215,7 @@ def dist_train():
elif
rank
<
15
:
parameter_server
()
else
:
optimiz
er
.
train
(
model
,
reader
=
read
)
train
er
.
train
(
model
,
reader
=
read
)
```
Please be aware that if a process is running on the Kubernetes
...
...
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