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
Crayon鑫
Paddle
提交
c3b009a0
P
Paddle
项目概览
Crayon鑫
/
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
提交
c3b009a0
编写于
11月 16, 2018
作者:
P
peizhilin
浏览文件
操作
浏览文件
下载
差异文件
Merge branch 'windows/build' into windows/online
test=develop
上级
dc80be27
fcbd5a12
变更
1
隐藏空白更改
内联
并排
Showing
1 changed file
with
58 addition
and
60 deletion
+58
-60
python/paddle/fluid/layers/io.py
python/paddle/fluid/layers/io.py
+58
-60
未找到文件。
python/paddle/fluid/layers/io.py
浏览文件 @
c3b009a0
...
...
@@ -347,72 +347,70 @@ def _copy_reader_create_op_(block, op):
return
new_op
if
os
.
name
!=
'nt'
:
@
templatedoc
(
op_type
=
'create_recordio_file_reader'
)
def
open_recordio_file
(
filename
,
shapes
,
lod_levels
,
dtypes
,
pass_num
=
1
,
for_parallel
=
True
):
"""
${comment}
Args:
filename(${filename_type}): ${filename_comment}.
shapes(list): List of tuples which declaring data shapes.
lod_levels(${lod_levels_type}): ${lod_levels_comment}.
dtypes(list): List of strs which declaring data type.
pass_num(int): Number of passes to run.
for_parallel(Bool): Set it as True if you are going to run
subsequent operators in parallel.
Returns:
${out_comment}.
Examples:
>>> import paddle.fluid as fluid
>>> reader = fluid.layers.io.open_recordio_file(
>>> filename='./data.recordio',
>>> shapes=[(3,224,224), (1)],
>>> lod_levels=[0, 0],
>>> dtypes=['float32', 'int64'])
>>> # Via the reader, we can use 'read_file' layer to get data:
>>> image, label = fluid.layers.io.read_file(reader)
"""
dtypes
=
[
convert_np_dtype_to_dtype_
(
dt
)
for
dt
in
dtypes
]
shape_concat
=
[]
ranks
=
[]
@
templatedoc
(
op_type
=
'create_recordio_file_reader'
)
def
open_recordio_file
(
filename
,
shapes
,
lod_levels
,
dtypes
,
pass_num
=
1
,
for_parallel
=
True
):
"""
${comment}
for
shape
in
shapes
:
shape_concat
.
extend
(
shape
)
ranks
.
append
(
len
(
shape
))
Args:
filename(${filename_type}): ${filename_comment}.
shapes(list): List of tuples which declaring data shapes.
lod_levels(${lod_levels_type}): ${lod_levels_comment}.
dtypes(list): List of strs which declaring data type.
pass_num(int): Number of passes to run.
for_parallel(Bool): Set it as True if you are going to run
subsequent operators in parallel.
var_name
=
unique_name
(
'open_recordio_file'
)
Returns:
${out_comment}.
startup_blk
=
default_startup_program
().
current_block
()
startup_var
=
startup_blk
.
create_var
(
name
=
var_name
)
startup_blk
.
append_op
(
type
=
'create_recordio_file_reader'
,
outputs
=
{
'Out'
:
[
startup_var
]},
attrs
=
{
'shape_concat'
:
shape_concat
,
'lod_levels'
:
lod_levels
,
'filename'
:
filename
,
'ranks'
:
ranks
})
Examples:
startup_var
.
desc
.
set_dtypes
(
dtypes
)
startup_var
.
persistable
=
True
main_prog_var
=
_copy_reader_var_
(
default_main_program
().
current_block
(),
startup_var
)
>>> import paddle.fluid as fluid
>>> reader = fluid.layers.io.open_recordio_file(
>>> filename='./data.recordio',
>>> shapes=[(3,224,224), (1)],
>>> lod_levels=[0, 0],
>>> dtypes=['float32', 'int64'])
>>> # Via the reader, we can use 'read_file' layer to get data:
>>> image, label = fluid.layers.io.read_file(reader)
"""
dtypes
=
[
convert_np_dtype_to_dtype_
(
dt
)
for
dt
in
dtypes
]
shape_concat
=
[]
ranks
=
[]
for
shape
in
shapes
:
shape_concat
.
extend
(
shape
)
ranks
.
append
(
len
(
shape
))
if
pass_num
>
1
:
main_prog_var
=
multi_pass
(
reader
=
main_prog_var
,
pass_num
=
pass_num
)
var_name
=
unique_name
(
'open_recordio_file'
)
return
monkey_patch_reader_methods
(
main_prog_var
)
startup_blk
=
default_startup_program
().
current_block
()
startup_var
=
startup_blk
.
create_var
(
name
=
var_name
)
startup_blk
.
append_op
(
type
=
'create_recordio_file_reader'
,
outputs
=
{
'Out'
:
[
startup_var
]},
attrs
=
{
'shape_concat'
:
shape_concat
,
'lod_levels'
:
lod_levels
,
'filename'
:
filename
,
'ranks'
:
ranks
})
startup_var
.
desc
.
set_dtypes
(
dtypes
)
startup_var
.
persistable
=
True
main_prog_var
=
_copy_reader_var_
(
default_main_program
().
current_block
(),
startup_var
)
if
pass_num
>
1
:
main_prog_var
=
multi_pass
(
reader
=
main_prog_var
,
pass_num
=
pass_num
)
return
monkey_patch_reader_methods
(
main_prog_var
)
def
random_data_generator
(
low
,
high
,
shapes
,
lod_levels
,
for_parallel
=
True
):
...
...
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