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
s920243400
PaddleDetection
提交
7ca85280
P
PaddleDetection
项目概览
s920243400
/
PaddleDetection
与 Fork 源项目一致
Fork自
PaddlePaddle / PaddleDetection
通知
2
Star
0
Fork
0
代码
文件
提交
分支
Tags
贡献者
分支图
Diff
Issue
0
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
PaddleDetection
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
0
Issue
0
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
未验证
提交
7ca85280
编写于
7月 13, 2018
作者:
Y
yuyang18
浏览文件
操作
浏览文件
下载
电子邮件补丁
差异文件
Refine pyreader demo
上级
e8eb81ca
变更
2
隐藏空白更改
内联
并排
Showing
2 changed file
with
18 addition
and
5 deletion
+18
-5
python/paddle/fluid/layers/io.py
python/paddle/fluid/layers/io.py
+13
-2
python/paddle/fluid/tests/demo/pyreader.py
python/paddle/fluid/tests/demo/pyreader.py
+5
-3
未找到文件。
python/paddle/fluid/layers/io.py
浏览文件 @
7ca85280
...
...
@@ -447,7 +447,12 @@ def random_data_generator(low, high, shapes, lod_levels, for_parallel=True):
return
monkey_patch_reader_methods
(
main_prog_var
)
def
py_reader
(
capacity
,
shapes
,
dtypes
,
lod_levels
=
None
,
name
=
None
):
def
py_reader
(
capacity
,
shapes
,
dtypes
,
lod_levels
=
None
,
name
=
None
,
use_double_buffer
=
True
):
"""
Create a reader and blocking queue for data feeding in Python
...
...
@@ -460,6 +465,7 @@ def py_reader(capacity, shapes, dtypes, lod_levels=None, name=None):
using `close()` method when unused.
Args:
use_double_buffer(bool): Whether use double buffer or not.
capacity(int): The maximum capacity of the BlockingQueue.
shapes(list|tuple): List of tuples which declaring data shapes.
dtypes(list|tuple): List of strs which declaring data type.
...
...
@@ -509,9 +515,11 @@ def py_reader(capacity, shapes, dtypes, lod_levels=None, name=None):
if
name
is
None
:
queue_name
=
unique_name
(
'lod_tensor_blocking_queue'
)
reader_name
=
unique_name
(
'create_py_reader'
)
double_buffer_name
=
unique_name
(
'double_buffer'
)
else
:
queue_name
=
"_"
.
join
([
name
,
"queue"
])
reader_name
=
"_"
.
join
([
name
,
"reader"
])
double_buffer_name
=
"_"
.
join
([
name
,
"double_buffer"
])
var
=
global_scope
().
var
(
queue_name
)
feed_queue
=
core
.
init_lod_tensor_blocking_queue
(
var
,
capacity
,
shapes
)
...
...
@@ -534,7 +542,10 @@ def py_reader(capacity, shapes, dtypes, lod_levels=None, name=None):
main_prog_var
=
_copy_reader_var_
(
default_main_program
().
current_block
(),
startup_var
)
return
monkey_patch_reader_methods
(
main_prog_var
),
feed_queue
reader
=
monkey_patch_reader_methods
(
main_prog_var
)
if
use_double_buffer
:
reader
=
double_buffer
(
reader
,
name
=
double_buffer_name
)
return
reader
,
feed_queue
def
open_files
(
filenames
,
...
...
python/paddle/fluid/tests/demo/pyreader.py
浏览文件 @
7ca85280
...
...
@@ -26,7 +26,7 @@ def network(is_train):
shapes
=
((
-
1
,
784
),
(
-
1
,
1
)),
dtypes
=
(
'float32'
,
'int64'
),
name
=
"train_reader"
if
is_train
else
"test_reader"
)
img
,
label
=
fluid
.
layers
.
read_file
(
fluid
.
layers
.
double_buffer
(
reader
)
)
img
,
label
=
fluid
.
layers
.
read_file
(
reader
)
hidden
=
img
...
...
@@ -100,7 +100,7 @@ def main():
trainer
.
run
(
fetch_list
=
[
loss
.
name
]))
except
fluid
.
core
.
EOFException
:
print
'End of epoch'
,
epoch_id
train_reader
.
reset
()
#
train_reader.reset()
train_data_thread
.
join
()
test_data_thread
=
pipe_reader_to_queue
(
...
...
@@ -111,10 +111,12 @@ def main():
tester
.
run
(
fetch_list
=
[
test_loss
.
name
]))
except
fluid
.
core
.
EOFException
:
print
'End of testing'
test_reader
.
reset
()
#
test_reader.reset()
test_data_thread
.
join
()
break
del
trainer
del
tester
if
__name__
==
'__main__'
:
...
...
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