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
8ab22a0c
提交
8ab22a0c
authored
5月 21, 2015
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
typo and opt name definition
上级
640b4347
隐藏空白字符变更
内嵌
并排
正在显示
1 个修改的文件
包含
10 行增加
和
3 行删除
+10
-3
optimization.txt
doc/extending/optimization.txt
+10
-3
没有找到文件。
doc/extending/optimization.txt
浏览文件 @
8ab22a0c
...
...
@@ -962,7 +962,7 @@ To understand this profile here is some explanation of how optimization work:
* ``0.751816s - ('canonicalize', 'EquilibriumOptimizer', 4) - 0.004s``
This line is from SeqOptimizer. Is mean that this sub optimizer took
a total of .7s. Its name is canonicalize. It is an
'EquilibriumOptimizer'. It was execute
r
at index 4 by the
'EquilibriumOptimizer'. It was execute
d
at index 4 by the
SeqOptimizer. It spent 0.004s in the validate phase.
* All other lines are from the profiler of the EquilibriumOptimizer.
...
...
@@ -978,11 +978,11 @@ To understand this profile here is some explanation of how optimization work:
the graph had 108 node, at the end, it had 81 and the maximum size
was 177.
* Then it print some global tim
m
ing, like is spent 0.029s in
* Then it print some global timing, like is spent 0.029s in
io_toposort, all local optimizer took 0.687s together for all
passes and global optimizer took a total of 0.010s.
* Then we pri
t the tim
ming for each pass and the optimization that
* Then we pri
nt the ti
ming for each pass and the optimization that
got applied and the number of time they got applied. For example,
in pass 0, the local_dimshuffle_lift optimizer changed the graph 9
time.
...
...
@@ -990,3 +990,10 @@ To understand this profile here is some explanation of how optimization work:
* Then we print the time spend in each optimizer, the number of time
they changed the graph and the number of node they introduced in
the graph.
* Optimization with that pattern `local_op_lift` mean that a node
with that op will be replaced by another node, with the same op,
but will do computation closer to the inputs of the graph.
* Optimization with that pattern `local_op_sink` is the opposite of
`lift`.
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