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
7c59ac48
P
Paddle
项目概览
PaddlePaddle
/
Paddle
12 个月 前同步成功
通知
2292
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 >>
提交
7c59ac48
编写于
3月 20, 2018
作者:
W
wanghaoshuang
浏览文件
操作
浏览文件
下载
电子邮件补丁
差异文件
Refine doc and use 'raise' instead of assert
上级
de2d7299
变更
2
显示空白变更内容
内联
并排
Showing
2 changed file
with
43 addition
and
6 deletion
+43
-6
doc/v2/api/fluid/optimizer.rst
doc/v2/api/fluid/optimizer.rst
+7
-0
python/paddle/fluid/optimizer.py
python/paddle/fluid/optimizer.py
+36
-6
未找到文件。
doc/v2/api/fluid/optimizer.rst
浏览文件 @
7c59ac48
...
@@ -47,3 +47,10 @@ DecayedAdagrad
...
@@ -47,3 +47,10 @@ DecayedAdagrad
:members:
:members:
:noindex:
:noindex:
Adadelta
--------------
.. autoclass:: paddle.fluid.optimizer.AdadeltaOptimizer
:members:
:noindex:
python/paddle/fluid/optimizer.py
浏览文件 @
7c59ac48
...
@@ -583,16 +583,44 @@ class DecayedAdagradOptimizer(Optimizer):
...
@@ -583,16 +583,44 @@ class DecayedAdagradOptimizer(Optimizer):
class
AdadeltaOptimizer
(
Optimizer
):
class
AdadeltaOptimizer
(
Optimizer
):
"""Simple Adadelta optimizer with average squared grad state and
"""
**Adadelta Optimizer**
Simple Adadelta optimizer with average squared grad state and
average squared update state.
average squared update state.
The details of adadelta please refer to this
`ADADELTA: AN ADAPTIVE LEARNING RATE METHOD
<http://www.matthewzeiler.com/pubs/googleTR2012/googleTR2012.pdf>`_.
.. math::
E(g_t^2) &=
\\
rho * E(g_{t-1}^2) + (1-
\\
rho) * g^2
\\\\
learning
\\
_rate &= sqrt( ( E(dx_{t-1}^2) +
\\
epsilon ) / (
\\
E(g_t^2) +
\\
epsilon ) )
\\\\
E(dx_t^2) &=
\\
rho * E(dx_{t-1}^2) + (1-
\\
rho) * (-g*learning
\\
_rate)^2
Args:
learning_rate(float): global leraning rate
rho(float): rho in equation
epsilon(float): epsilon in equation
Examples:
.. code-block:: python
optimizer = fluid.optimizer.Adadelta(
learning_rate=0.0003, epsilon=1.0e-6, rho=0.95)
_, params_grads = optimizer.minimize(cost)
"""
"""
_avg_squared_grad_acc_str
=
"_avg_squared_grad"
_avg_squared_grad_acc_str
=
"_avg_squared_grad"
_avg_squared_update_acc_str
=
"_avg_squared_update"
_avg_squared_update_acc_str
=
"_avg_squared_update"
def
__init__
(
self
,
learning_rate
,
epsilon
=
1.0e-6
,
rho
=
0.95
,
**
kwargs
):
def
__init__
(
self
,
learning_rate
,
epsilon
=
1.0e-6
,
rho
=
0.95
,
**
kwargs
):
assert
learning_rate
is
not
None
if
learning_rate
is
None
:
assert
epsilon
is
not
None
raise
ValueError
(
"learning_rate is not set."
)
assert
rho
is
not
None
if
epsilon
is
None
:
raise
ValueError
(
"epsilon is not set."
)
if
rho
is
None
:
raise
ValueError
(
"rho is not set."
)
super
(
AdadeltaOptimizer
,
self
).
__init__
(
super
(
AdadeltaOptimizer
,
self
).
__init__
(
learning_rate
=
learning_rate
,
**
kwargs
)
learning_rate
=
learning_rate
,
**
kwargs
)
self
.
type
=
"adadelta"
self
.
type
=
"adadelta"
...
@@ -600,14 +628,16 @@ class AdadeltaOptimizer(Optimizer):
...
@@ -600,14 +628,16 @@ class AdadeltaOptimizer(Optimizer):
self
.
_rho
=
rho
self
.
_rho
=
rho
def
_create_accumulators
(
self
,
block
,
parameters
):
def
_create_accumulators
(
self
,
block
,
parameters
):
assert
isinstance
(
block
,
framework
.
Block
)
if
not
isinstance
(
block
,
framework
.
Block
):
raise
TypeError
(
"block is not instance of framework.Block."
)
for
p
in
parameters
:
for
p
in
parameters
:
self
.
_add_accumulator
(
self
.
_avg_squared_grad_acc_str
,
p
)
self
.
_add_accumulator
(
self
.
_avg_squared_grad_acc_str
,
p
)
self
.
_add_accumulator
(
self
.
_avg_squared_update_acc_str
,
p
)
self
.
_add_accumulator
(
self
.
_avg_squared_update_acc_str
,
p
)
def
_append_optimize_op
(
self
,
block
,
param_and_grad
):
def
_append_optimize_op
(
self
,
block
,
param_and_grad
):
assert
isinstance
(
block
,
framework
.
Block
)
if
not
isinstance
(
block
,
framework
.
Block
):
raise
TypeError
(
"block is not instance of framework.Block."
)
avg_squared_grad_acc
=
self
.
_get_accumulator
(
avg_squared_grad_acc
=
self
.
_get_accumulator
(
self
.
_avg_squared_grad_acc_str
,
param_and_grad
[
0
])
self
.
_avg_squared_grad_acc_str
,
param_and_grad
[
0
])
...
...
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