Skip to content
项目
群组
代码片段
帮助
当前项目
正在载入...
登录 / 注册
切换导航面板
P
pytensor
项目
项目
详情
活动
周期分析
仓库
仓库
文件
提交
分支
标签
贡献者
图表
比较
统计图
议题
0
议题
0
列表
看板
标记
里程碑
合并请求
0
合并请求
0
CI / CD
CI / CD
流水线
作业
日程
统计图
Wiki
Wiki
代码片段
代码片段
成员
成员
折叠边栏
关闭边栏
活动
图像
聊天
创建新问题
作业
提交
问题看板
Open sidebar
testgroup
pytensor
Commits
7c71f3ad
提交
7c71f3ad
authored
11月 18, 2014
作者:
Frederic
浏览文件
操作
浏览文件
下载
电子邮件补丁
差异文件
fix typo in doc/docstring
上级
461b7e02
隐藏空白字符变更
内嵌
并排
正在显示
2 个修改的文件
包含
10 行增加
和
10 行删除
+10
-10
faq.txt
doc/faq.txt
+9
-9
opt.py
theano/tensor/opt.py
+1
-1
没有找到文件。
doc/faq.txt
浏览文件 @
7c71f3ad
...
@@ -85,17 +85,17 @@ Unsafe optimization
...
@@ -85,17 +85,17 @@ Unsafe optimization
===================
===================
Some Theano optimizations make the assumption that the user inputs
is
Some Theano optimizations make the assumption that the user inputs
are
valid. What th
at mean is that if the user provide
invalid values (like
valid. What th
is means is that if the user provides
invalid values (like
not compatible shapes or indexing values that are out of bound
) and
incompatible shapes or indexing values that are out of bounds
) and
the optimizations
is applied, the user error will get lost.
Most of the
the optimizations
are applied, the user error will get lost.
Most of the
time, the assumption is that the user inputs are valid. So it is good
time, the assumption is that the user inputs are valid. So it is good
to have the optimization being applied
. B
ut loosing the error is bad.
to have the optimization being applied
, b
ut loosing the error is bad.
The newest optimization in Theano with such assumption will add
The newest optimization in Theano with such assumption will add
an
assertion in the graph to keep the user error message.
But c
omputing
assertion in the graph to keep the user error message.
C
omputing
th
ose assumption could take some time. If you are sure you all
is valid
th
ese assertions could take some time. If you are sure everything
is valid
in your graph and want the fastest possible Theano, you can enable an
in your graph and want the fastest possible Theano, you can enable an
optimization that will remove those assertion:
optimization that will remove those assertion
s with
:
``optimizer_including=local_remove_all_assert``
``optimizer_including=local_remove_all_assert``
...
...
theano/tensor/opt.py
浏览文件 @
7c71f3ad
...
@@ -1584,7 +1584,7 @@ def local_remove_useless_assert(node):
...
@@ -1584,7 +1584,7 @@ def local_remove_useless_assert(node):
@gof.local_optimizer
([
Assert
])
@gof.local_optimizer
([
Assert
])
def
local_remove_all_assert
(
node
):
def
local_remove_all_assert
(
node
):
"""An optimization disable
by default that remove all assert
from
"""An optimization disable
d by default that removes all asserts
from
the graph.
the graph.
:note: See the :ref:`unsafe` section to know how to enable it.
:note: See the :ref:`unsafe` section to know how to enable it.
...
...
编写
预览
Markdown
格式
0%
重试
或
添加新文件
添加附件
取消
您添加了
0
人
到此讨论。请谨慎行事。
请先完成此评论的编辑!
取消
请
注册
或者
登录
后发表评论