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
762cb0d1
提交
762cb0d1
authored
7月 22, 2015
作者:
Iban Harlouchet
提交者:
Arnaud Bergeron
9月 08, 2015
浏览文件
操作
浏览文件
下载
电子邮件补丁
差异文件
testcode for doc/tutorial/conditions.txt
上级
af235f4e
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
35 行增加
和
29 行删除
+35
-29
conditions.txt
doc/tutorial/conditions.txt
+35
-29
没有找到文件。
doc/tutorial/conditions.txt
浏览文件 @
762cb0d1
...
@@ -18,49 +18,55 @@ IfElse vs Switch
...
@@ -18,49 +18,55 @@ IfElse vs Switch
**Example**
**Example**
..
code-block:: python
..
testcode::
from theano import tensor as T
from theano import tensor as T
from theano.ifelse import ifelse
from theano.ifelse import ifelse
import theano, time, numpy
import theano, time, numpy
a,b = T.scalars('a', 'b')
a,b = T.scalars('a', 'b')
x,y = T.matrices('x', 'y')
x,y = T.matrices('x', 'y')
z_switch = T.switch(T.lt(a, b), T.mean(x), T.mean(y))
z_switch = T.switch(T.lt(a, b), T.mean(x), T.mean(y))
z_lazy = ifelse(T.lt(a, b), T.mean(x), T.mean(y))
z_lazy = ifelse(T.lt(a, b), T.mean(x), T.mean(y))
f_switch = theano.function([a, b, x, y], z_switch,
f_switch = theano.function([a, b, x, y], z_switch,
mode=theano.Mode(linker='vm'))
mode=theano.Mode(linker='vm'))
f_lazyifelse = theano.function([a, b, x, y], z_lazy,
f_lazyifelse = theano.function([a, b, x, y], z_lazy,
mode=theano.Mode(linker='vm'))
mode=theano.Mode(linker='vm'))
val1 = 0.
val1 = 0.
val2 = 1.
val2 = 1.
big_mat1 = numpy.ones((10000, 1000))
big_mat1 = numpy.ones((10000, 1000))
big_mat2 = numpy.ones((10000, 1000))
big_mat2 = numpy.ones((10000, 1000))
n_times = 10
n_times = 10
tic = time.clock()
tic = time.clock()
for i in xrange(n_times):
for i in xrange(n_times):
f_switch(val1, val2, big_mat1, big_mat2)
f_switch(val1, val2, big_mat1, big_mat2)
print 'time spent evaluating both values %f sec' % (time.clock() - tic)
print 'time spent evaluating both values %f sec' % (time.clock() - tic)
tic = time.clock()
tic = time.clock()
for i in xrange(n_times):
for i in xrange(n_times):
f_lazyifelse(val1, val2, big_mat1, big_mat2)
f_lazyifelse(val1, val2, big_mat1, big_mat2)
print 'time spent evaluating one value %f sec' % (time.clock() - tic)
print 'time spent evaluating one value %f sec' % (time.clock() - tic)
In this example, the ``IfElse`` op spends less time (about half as much) than ``Switch``
In this example, the ``IfElse`` op spends less time (about half as much) than ``Switch``
since it computes only one variable out of the two.
since it computes only one variable out of the two.
.. code-block:: python
.. testoutput::
:hide:
:options: +ELLIPSIS
>>> python ifelse_switch.py
time spent evaluating both values ... sec
time spent evaluating both values 0.6700 sec
time spent evaluating one value ... sec
time spent evaluating one value 0.3500 sec
.. code-block:: none
$ python ifelse_switch.py
time spent evaluating both values 0.6700 sec
time spent evaluating one value 0.3500 sec
Unless ``linker='vm'`` or ``linker='cvm'`` are used, ``ifelse`` will compute both
Unless ``linker='vm'`` or ``linker='cvm'`` are used, ``ifelse`` will compute both
variables and take the same computation time as ``switch``. Although the linker
variables and take the same computation time as ``switch``. Although the linker
...
...
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