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
jobily
PaddleOCR
提交
d3c50fda
P
PaddleOCR
项目概览
jobily
/
PaddleOCR
与 Fork 源项目一致
Fork自
PaddlePaddle / PaddleOCR
通知
1
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
PaddleOCR
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
体验新版 GitCode,发现更多精彩内容 >>
提交
d3c50fda
编写于
1月 20, 2021
作者:
只会git clone的程序员
浏览文件
操作
浏览文件
下载
电子邮件补丁
差异文件
fix bugs
上级
46ac85ad
变更
3
隐藏空白更改
内联
并排
Showing
3 changed file
with
36 addition
and
14 deletion
+36
-14
configs/rec/multi_language/generate_multi_language_configs.py
...igs/rec/multi_language/generate_multi_language_configs.py
+34
-12
configs/rec/multi_language/rec_multi_language_lite_train.yml
configs/rec/multi_language/rec_multi_language_lite_train.yml
+2
-2
imgs.zip
imgs.zip
+0
-0
未找到文件。
configs/rec/multi_language/generate_multi_language_configs.py
浏览文件 @
d3c50fda
import
yaml
from
argparse
import
ArgumentParser
,
RawDescriptionHelpFormatter
import
os.path
import
logging
logging
.
basicConfig
(
level
=
logging
.
INFO
)
support_list
=
{
'it'
:
'italian'
,
'xi'
:
'spanish'
,
'pu'
:
'portuguese'
,
'ru'
:
'russian'
,
'ar'
:
'arabic'
,
...
...
@@ -16,6 +18,7 @@ You can download it from \
https://github.com/PaddlePaddle/PaddleOCR/tree/dygraph/configs/rec/multi_language/"
global_config
=
yaml
.
load
(
open
(
"./rec_multi_language_lite_train.yml"
,
'rb'
),
Loader
=
yaml
.
Loader
)
project_path
=
os
.
path
.
abspath
(
os
.
path
.
join
(
os
.
getcwd
(),
"../../../"
))
class
ArgsParser
(
ArgumentParser
):
def
__init__
(
self
):
...
...
@@ -32,7 +35,7 @@ class ArgsParser(ArgumentParser):
self
.
add_argument
(
"--dict"
,
type
=
str
,
help
=
"you can use this command to change the dictionary default path"
)
self
.
add_argument
(
"--data
set_root_path
"
,
type
=
str
,
help
=
"you can use this command to change the dataset default root path"
)
"--data
_dir
"
,
type
=
str
,
help
=
"you can use this command to change the dataset default root path"
)
def
parse_args
(
self
,
argv
=
None
):
args
=
super
(
ArgsParser
,
self
).
parse_args
(
argv
)
...
...
@@ -51,15 +54,19 @@ class ArgsParser(ArgumentParser):
return
config
def
_set_language
(
self
,
type
):
assert
(
type
),
"please use -
t or --typ
e to choose language type"
assert
(
type
),
"please use -
l or --languag
e to choose language type"
assert
(
type
[
0
]
in
support_list
.
keys
()
),
"the sub_keys(-
t or --typ
e) can only be one of support list:
\n
{},
\n
but get: {}, "
\
),
"the sub_keys(-
l or --languag
e) can only be one of support list:
\n
{},
\n
but get: {}, "
\
"please check your running command"
.
format
(
support_list
,
type
)
global_config
[
'Global'
][
'character_dict_path'
]
=
'ppocr/utils/dict/{}_dict.txt'
.
format
(
type
[
0
])
global_config
[
'Global'
][
'save_model_dir'
]
=
'./output/rec_{}_lite'
.
format
(
type
[
0
])
global_config
[
'Train'
][
'dataset'
][
'label_file_list'
]
=
[
"train_data/{}_train.txt"
.
format
(
type
[
0
])]
global_config
[
'Eval'
][
'dataset'
][
'label_file_list'
]
=
[
"train_data/{}_val.txt"
.
format
(
type
[
0
])]
assert
(
os
.
path
.
isfile
(
os
.
path
.
join
(
project_path
,
global_config
[
'Global'
][
'character_dict_path'
]))
),
"Loss default dictionary file {}_dict.txt.You can download it from
\
https://github.com/PaddlePaddle/PaddleOCR/tree/dygraph/ppocr/utils/dict/"
.
format
(
type
[
0
])
return
type
[
0
]
...
...
@@ -88,27 +95,42 @@ def merge_config(config):
cur
[
sub_key
]
=
value
else
:
cur
=
cur
[
sub_key
]
def
loss_file
(
path
):
if
not
os
.
path
.
exists
(
path
):
logging
.
warning
(
'There is no such file:{},Please do not forget to put in the specified file'
.
format
(
path
))
if
__name__
==
'__main__'
:
FLAGS
=
ArgsParser
().
parse_args
()
merge_config
(
FLAGS
.
opt
)
if
FLAGS
.
train
:
global_config
[
'Train'
][
'dataset'
][
'label_file_list'
]
=
[
FLAGS
.
train
]
train_label_path
=
os
.
path
.
join
(
project_path
,
FLAGS
.
train
)
loss_file
(
train_label_path
)
if
FLAGS
.
val
:
global_config
[
'Eval'
][
'dataset'
][
'label_file_list'
]
=
[
FLAGS
.
val
]
eval_label_path
=
os
.
path
.
join
(
project_path
,
FLAGS
.
val
)
loss_file
(
Eval_label_path
)
if
FLAGS
.
dict
:
global_config
[
'Global'
][
'character_dict_path'
]
=
FLAGS
.
dict
if
FLAGS
.
dataset_root_path
:
global_config
[
'Eval'
][
'dataset'
][
'data_dir'
]
=
FLAGS
.
dataset_root_path
global_config
[
'Train'
][
'dataset'
][
'data_dir'
]
=
FLAGS
.
dataset_root_path
dict_path
=
os
.
path
.
join
(
project_path
,
FLAGS
.
dict
)
loss_file
(
dict_path
)
if
FLAGS
.
data_dir
:
global_config
[
'Eval'
][
'dataset'
][
'data_dir'
]
=
FLAGS
.
data_dir
global_config
[
'Train'
][
'dataset'
][
'data_dir'
]
=
FLAGS
.
data_dir
data_dir
=
os
.
path
.
join
(
project_path
,
FLAGS
.
data_dir
)
loss_file
(
data_dir
)
save_file_path
=
'rec_{}_lite_train.yml'
.
format
(
FLAGS
.
language
)
if
os
.
path
.
isfile
(
save_file_path
):
os
.
remove
(
save_file_path
)
with
open
(
save_file_path
,
'w'
)
as
f
:
yaml
.
dump
(
dict
(
global_config
),
f
,
default_flow_style
=
False
,
sort_keys
=
False
)
print
(
"Train list path set to :{}"
.
format
(
global_config
[
'Train'
][
'dataset'
][
'label_file_list'
][
0
]))
print
(
"Eval list path set to :{}"
.
format
(
global_config
[
'Eval'
][
'dataset'
][
'label_file_list'
][
0
]))
print
(
"Dataset root path set to :{}"
.
format
(
global_config
[
'Eval'
][
'dataset'
][
'data_dir'
]))
print
(
"Dict path set to :{}"
.
format
(
global_config
[
'Global'
][
'character_dict_path'
]))
print
(
"Config file set to :configs/rec/multi_language/{}"
.
format
(
save_file_path
))
logging
.
info
(
"Project path is :{}"
.
format
(
project_path
))
logging
.
info
(
"Train list path set to :{}"
.
format
(
global_config
[
'Train'
][
'dataset'
][
'label_file_list'
][
0
]))
logging
.
info
(
"Eval list path set to :{}"
.
format
(
global_config
[
'Eval'
][
'dataset'
][
'label_file_list'
][
0
]))
logging
.
info
(
"Dataset root path set to :{}"
.
format
(
global_config
[
'Eval'
][
'dataset'
][
'data_dir'
]))
logging
.
info
(
"Dict path set to :{}"
.
format
(
global_config
[
'Global'
][
'character_dict_path'
]))
logging
.
info
(
"Config file set to :configs/rec/multi_language/{}"
.
format
(
save_file_path
))
configs/rec/multi_language/rec_multi_language_lite_train.yml
浏览文件 @
d3c50fda
...
...
@@ -64,7 +64,7 @@ Metric:
Train
:
dataset
:
name
:
SimpleDataSet
data_dir
:
./
train_data/
data_dir
:
train_data/
label_file_list
:
[
"
./train_data/train_list.txt"
]
transforms
:
-
DecodeImage
:
# load image
...
...
@@ -85,7 +85,7 @@ Train:
Eval
:
dataset
:
name
:
SimpleDataSet
data_dir
:
./
train_data/
data_dir
:
train_data/
label_file_list
:
[
"
./train_data/val_list.txt"
]
transforms
:
-
DecodeImage
:
# load image
...
...
imgs.zip
0 → 100644
浏览文件 @
d3c50fda
文件已添加
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