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
PaddlePaddle
Paddle
提交
e081c7a0
P
Paddle
项目概览
PaddlePaddle
/
Paddle
大约 1 年 前同步成功
通知
2298
Star
20931
Fork
5422
代码
文件
提交
分支
Tags
贡献者
分支图
Diff
Issue
1423
列表
看板
标记
里程碑
合并请求
543
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
1,423
Issue
1,423
列表
看板
标记
里程碑
合并请求
543
合并请求
543
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
未验证
提交
e081c7a0
编写于
3月 10, 2020
作者:
C
Chen Weihang
提交者:
GitHub
3月 10, 2020
浏览文件
操作
浏览文件
下载
电子邮件补丁
差异文件
skip type/dtype check in dygraph mode, test=develop (#22915)
上级
c5cbe7f0
变更
1
显示空白变更内容
内联
并排
Showing
1 changed file
with
13 addition
and
1 deletion
+13
-1
python/paddle/fluid/data_feeder.py
python/paddle/fluid/data_feeder.py
+13
-1
未找到文件。
python/paddle/fluid/data_feeder.py
浏览文件 @
e081c7a0
...
@@ -22,7 +22,7 @@ from six.moves import zip, range, xrange
...
@@ -22,7 +22,7 @@ from six.moves import zip, range, xrange
import
multiprocessing
import
multiprocessing
import
warnings
import
warnings
from
.framework
import
Variable
,
default_main_program
,
_current_expected_place
from
.framework
import
Variable
,
default_main_program
,
_current_expected_place
,
in_dygraph_mode
from
.framework
import
_cpu_num
,
_cuda_ids
from
.framework
import
_cpu_num
,
_cuda_ids
__all__
=
[
'DataFeeder'
]
__all__
=
[
'DataFeeder'
]
...
@@ -81,6 +81,15 @@ def check_variable_and_dtype(input,
...
@@ -81,6 +81,15 @@ def check_variable_and_dtype(input,
def
check_type
(
input
,
input_name
,
expected_type
,
op_name
,
extra_message
=
''
):
def
check_type
(
input
,
input_name
,
expected_type
,
op_name
,
extra_message
=
''
):
# NOTE [ Why skip dynamic graph check ]:
# 1. If the input type / dtype of a layer is wrong, it will be reported
# directly on that line. User can easily print the relevant information
# on which line. It is easier to debug, so there is no need to check
# in dynamic graph mode.
# 2. Performance considerations. Because these checks are executed at
# each step in dynamic graph mode, it will bring a heavy performance burden.
if
in_dygraph_mode
():
return
if
not
isinstance
(
input
,
expected_type
):
if
not
isinstance
(
input
,
expected_type
):
raise
TypeError
(
raise
TypeError
(
"The type of '%s' in %s must be %s, but received %s. %s"
%
"The type of '%s' in %s must be %s, but received %s. %s"
%
...
@@ -92,6 +101,9 @@ def check_dtype(input_dtype,
...
@@ -92,6 +101,9 @@ def check_dtype(input_dtype,
expected_dtype
,
expected_dtype
,
op_name
,
op_name
,
extra_message
=
''
):
extra_message
=
''
):
# See NOTE [ Why skip dynamic graph check ]
if
in_dygraph_mode
():
return
if
convert_dtype
(
input_dtype
)
in
[
'float16'
]:
if
convert_dtype
(
input_dtype
)
in
[
'float16'
]:
warnings
.
warn
(
warnings
.
warn
(
"The data type of '%s' in %s only support float16 in GPU now. %s"
%
"The data type of '%s' in %s only support float16 in GPU now. %s"
%
...
...
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