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
OPTHREE
Paddle
提交
706a7897
P
Paddle
项目概览
OPTHREE
/
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 >>
未验证
提交
706a7897
编写于
11月 18, 2021
作者:
Z
zhangbo9674
提交者:
GitHub
11月 18, 2021
浏览文件
操作
浏览文件
下载
电子邮件补丁
差异文件
Fix Layer.to() of device bug (#37156)
上级
34a44d59
变更
1
隐藏空白更改
内联
并排
Showing
1 changed file
with
12 addition
and
22 deletion
+12
-22
python/paddle/fluid/dygraph/layers.py
python/paddle/fluid/dygraph/layers.py
+12
-22
未找到文件。
python/paddle/fluid/dygraph/layers.py
浏览文件 @
706a7897
...
...
@@ -1556,19 +1556,18 @@ class Layer(core.Layer):
if
dtype
is
None
:
dtype
=
t
.
dtype
if
type
(
dtype
)
is
str
:
dtype
=
convert_np_dtype_to_dtype_
(
dtype
)
# 1. gpu place need to determine whether the memory is sufficient for allocation:
if
t
.
place
.
is_gpu_place
():
gpu_memory_available
=
core
.
gpu_memory_available
()
# for gpu, minimum memory allocation unit is 256 bytes.
if
type
(
dtype
)
is
str
:
size_dtype
=
core
.
size_of_dtype
(
convert_np_dtype_to_dtype_
(
dtype
))
else
:
size_dtype
=
core
.
size_of_dtype
(
dtype
)
size_dtype
=
core
.
size_of_dtype
(
dtype
)
# Note(zhangbo): Paddle GPU minimum memory allocation unit is 256 bytes, waiting_alloc_memory will comput ‘t’ occupied memory space.
# Coefficient 1.2 is used to avoid OOM that may occur in this critical state when the memory is just enough.
waiting_alloc_memory
=
(
(
t
.
numel
().
numpy
()[
0
]
*
size_dtype
)
/
256
+
1
)
*
256
*
1.2
(
np
.
prod
(
t
.
shape
)
*
size_dtype
)
/
256
+
1
)
*
256
*
1.2
gpu_memory_available
=
core
.
gpu_memory_available
()
if
gpu_memory_available
<
waiting_alloc_memory
:
# Copy param / Tensor to cpu
t_used
=
t
.
_copy_to
(
paddle
.
CPUPlace
(),
...
...
@@ -1582,26 +1581,17 @@ class Layer(core.Layer):
# 2. cast param / Tensor to dtype
if
dtype
is
not
None
and
dtype
!=
t_used
.
dtype
:
if
isinstance
(
t_used
,
framework
.
ParamBase
):
from
paddle.fluid.layer_helper
import
LayerHelper
helper
=
LayerHelper
(
"cast"
,
**
locals
())
t_casted
=
helper
.
create_variable_for_type_inference
(
dtype
=
dtype
)
framework
.
_dygraph_tracer
().
trace_op
(
type
=
'cast'
,
inputs
=
{
'X'
:
t_used
},
outputs
=
{
'Out'
:
t_casted
},
attrs
=
{
'in_dtype'
:
t_used
.
dtype
,
'out_dtype'
:
convert_np_dtype_to_dtype_
(
dtype
)
})
else
:
with
paddle
.
fluid
.
framework
.
_dygraph_place_guard
(
place
=
t_used
.
place
):
t_casted
=
t_used
.
cast
(
dtype
=
dtype
)
else
:
t_casted
=
t_used
# 3. Copy casted cpu param / Tensor to device
new_t
=
t_casted
.
_copy_to
(
device
,
blocking
)
if
device
is
not
None
and
not
t_casted
.
place
.
_equals
(
device
):
new_t
=
t_casted
.
_copy_to
(
device
,
blocking
)
else
:
new_t
=
t_casted
# 4. share Tensor to origin param / Tensor
dst_tensor
=
t
.
value
().
get_tensor
()
...
...
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