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
9499b886
提交
9499b886
authored
5月 20, 2014
作者:
Frédéric Bastien
浏览文件
操作
浏览文件
下载
差异文件
Merge pull request #1865 from caglar/fix_typos
fixed a few typos and removed unnecessary code.
上级
01f54ac2
480f45dc
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
2 行增加
和
4 行删除
+2
-4
opt.py
theano/sandbox/cuda/opt.py
+2
-4
没有找到文件。
theano/sandbox/cuda/opt.py
浏览文件 @
9499b886
...
@@ -88,7 +88,8 @@ register_opt(name='gpu_constant_folding')(
...
@@ -88,7 +88,8 @@ register_opt(name='gpu_constant_folding')(
class
InputToGpuOptimizer
(
Optimizer
):
class
InputToGpuOptimizer
(
Optimizer
):
"""Transfert the input of a graph to the gpu if needed
"""
Transfer the input of a graph to the gpu if it is necessary.
It should make this part of the optimizer faster we will will need only 1
It should make this part of the optimizer faster we will will need only 1
pass on the fgraph.
pass on the fgraph.
"""
"""
...
@@ -505,7 +506,6 @@ def local_gpu_gemv(node):
...
@@ -505,7 +506,6 @@ def local_gpu_gemv(node):
if
isinstance
(
node
.
op
,
GpuFromHost
):
if
isinstance
(
node
.
op
,
GpuFromHost
):
host_input
=
node
.
inputs
[
0
]
host_input
=
node
.
inputs
[
0
]
if
host_input
.
owner
and
isinstance
(
host_input
.
owner
.
op
,
gemvs
):
if
host_input
.
owner
and
isinstance
(
host_input
.
owner
.
op
,
gemvs
):
op
=
host_input
.
owner
.
op
z
,
a
,
x
,
y
,
b
=
host_input
.
owner
.
inputs
z
,
a
,
x
,
y
,
b
=
host_input
.
owner
.
inputs
return
[
gpu_gemv_no_inplace
(
return
[
gpu_gemv_no_inplace
(
gpu_from_host
(
z
),
gpu_from_host
(
z
),
...
@@ -546,7 +546,6 @@ def local_gpu_ger(node):
...
@@ -546,7 +546,6 @@ def local_gpu_ger(node):
if
isinstance
(
node
.
op
,
GpuFromHost
):
if
isinstance
(
node
.
op
,
GpuFromHost
):
host_input
=
node
.
inputs
[
0
]
host_input
=
node
.
inputs
[
0
]
if
host_input
.
owner
and
isinstance
(
host_input
.
owner
.
op
,
gers
):
if
host_input
.
owner
and
isinstance
(
host_input
.
owner
.
op
,
gers
):
op
=
host_input
.
owner
.
op
z
,
a
,
x
,
y
=
host_input
.
owner
.
inputs
z
,
a
,
x
,
y
=
host_input
.
owner
.
inputs
return
[
gpu_ger_no_inplace
(
return
[
gpu_ger_no_inplace
(
gpu_from_host
(
z
),
gpu_from_host
(
z
),
...
@@ -582,7 +581,6 @@ def local_gpu_gemm(node):
...
@@ -582,7 +581,6 @@ def local_gpu_gemm(node):
host_input
=
node
.
inputs
[
0
]
host_input
=
node
.
inputs
[
0
]
if
host_input
.
owner
and
isinstance
(
host_input
.
owner
.
op
,
if
host_input
.
owner
and
isinstance
(
host_input
.
owner
.
op
,
tensor
.
blas
.
Gemm
):
tensor
.
blas
.
Gemm
):
op
=
host_input
.
owner
.
op
z
,
a
,
x
,
y
,
b
=
host_input
.
owner
.
inputs
z
,
a
,
x
,
y
,
b
=
host_input
.
owner
.
inputs
return
[
gpu_gemm_no_inplace
(
gpu_from_host
(
z
),
return
[
gpu_gemm_no_inplace
(
gpu_from_host
(
z
),
a
,
a
,
...
...
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