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
溜达小灰鼠
Paddle
提交
6b9f1d34
P
Paddle
项目概览
溜达小灰鼠
/
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
1
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
1
Issue
1
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
6b9f1d34
编写于
1月 29, 2018
作者:
G
guosheng
浏览文件
操作
浏览文件
下载
电子邮件补丁
差异文件
Make weight normalization adapt to the up-to-date code
上级
a422f346
变更
3
隐藏空白更改
内联
并排
Showing
3 changed file
with
8 addition
and
5 deletion
+8
-5
python/paddle/v2/fluid/layer_helper.py
python/paddle/v2/fluid/layer_helper.py
+2
-2
python/paddle/v2/fluid/param_attr.py
python/paddle/v2/fluid/param_attr.py
+5
-2
python/paddle/v2/fluid/tests/test_weight_normalization.py
python/paddle/v2/fluid/tests/test_weight_normalization.py
+1
-1
未找到文件。
python/paddle/v2/fluid/layer_helper.py
浏览文件 @
6b9f1d34
...
...
@@ -226,8 +226,8 @@ class LayerHelper(object):
scale
=
elementwise_div
(
x
=
g
,
y
=
norm
)
# The shapes of g and norm are the same.
# Currently, elementwise_mul only support broadcast when the shape
# of y is a subset of
x. Thus, we should reshape y to squeeze to
#
achive i
t.
# of y is a subset of
the shape of x. Thus, we reshape y to squeeze
#
to achive the subse
t.
w
=
elementwise_mul
(
x
=
v
,
y
=
scale
if
dim
is
None
else
reshape
(
...
...
python/paddle/v2/fluid/param_attr.py
浏览文件 @
6b9f1d34
...
...
@@ -15,7 +15,10 @@
from
initializer
import
Initializer
,
Xavier
,
Constant
from
regularizer
import
WeightDecayRegularizer
__all__
=
[
'ParamAttr'
]
__all__
=
[
'ParamAttr'
,
'WeightNormParamAttr'
,
]
class
ParamAttr
(
object
):
...
...
@@ -92,7 +95,7 @@ class WeightNormParamAttr(ParamAttr):
"""
# List to record the parameters reparameterized by weight normalization.
# If these parameters are treated as Variable rather than Parameter,
# it can be used to discriminate these parameters and help to serialize
# it can be used to discriminate these parameters and help to serialize
# these paramters for inference.
params_with_weight_norm
=
[]
...
...
python/paddle/v2/fluid/tests/test_weight_normalization.py
浏览文件 @
6b9f1d34
...
...
@@ -52,7 +52,7 @@ class TestWeightNormalization(unittest.TestCase):
def
run_program
(
self
):
outputs
=
[]
places
=
[
core
.
CPUPlace
()]
if
core
.
is_compile
_gpu
():
if
core
.
is_compile
d_with_cuda
():
places
.
append
(
core
.
CUDAPlace
(
0
))
for
place
in
places
:
self
.
set_inputs
(
place
)
...
...
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