Merge pull request #1353 from tink2123/update_multi

update for multi-languages
release/2.0-rc1-0
xiaoting 4 years ago committed by GitHub
commit 0ecc335d9d
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23

@ -15,7 +15,7 @@ Global:
use_visualdl: False
infer_img:
# for data or label process
character_dict_path: ppocr/utils/ic15_dict.txt
character_dict_path: ppocr/utils/dict/ic15_dict.txt
character_type: ch
max_text_length: 25
infer_mode: False

@ -15,7 +15,7 @@ Global:
use_visualdl: False
infer_img:
# for data or label process
character_dict_path: ppocr/utils/french_dict.txt
character_dict_path: ppocr/utils/dict/french_dict.txt
character_type: french
max_text_length: 25
infer_mode: False

@ -15,7 +15,7 @@ Global:
use_visualdl: False
infer_img:
# for data or label process
character_dict_path: ppocr/utils/german_dict.txt
character_dict_path: ppocr/utils/dict/german_dict.txt
character_type: german
max_text_length: 25
infer_mode: False

@ -15,7 +15,7 @@ Global:
use_visualdl: False
infer_img:
# for data or label process
character_dict_path: ppocr/utils/japan_dict.txt
character_dict_path: ppocr/utils/dict/japan_dict.txt
character_type: japan
max_text_length: 25
infer_mode: False

@ -15,7 +15,7 @@ Global:
use_visualdl: False
infer_img:
# for data or label process
character_dict_path: ppocr/utils/korean_dict.txt
character_dict_path: ppocr/utils/dict/korean_dict.txt
character_type: korean
max_text_length: 25
infer_mode: False

@ -79,7 +79,9 @@ class BaseRecLabelEncode(object):
character_dict_path=None,
character_type='ch',
use_space_char=False):
support_character_type = ['ch', 'en', 'en_sensitive']
support_character_type = [
'ch', 'en', 'en_sensitive', 'french', 'german', 'japan', 'korean'
]
assert character_type in support_character_type, "Only {} are supported now but get {}".format(
support_character_type, self.character_str)
@ -87,7 +89,7 @@ class BaseRecLabelEncode(object):
if character_type == "en":
self.character_str = "0123456789abcdefghijklmnopqrstuvwxyz"
dict_character = list(self.character_str)
elif character_type == "ch":
elif character_type in ["ch", "french", "german", "japan", "korean"]:
self.character_str = ""
assert character_dict_path is not None, "character_dict_path should not be None when character_type is ch"
with open(character_dict_path, "rb") as fin:

@ -23,14 +23,16 @@ class BaseRecLabelDecode(object):
character_dict_path=None,
character_type='ch',
use_space_char=False):
support_character_type = ['ch', 'en', 'en_sensitive']
support_character_type = [
'ch', 'en', 'en_sensitive', 'french', 'german', 'japan', 'korean'
]
assert character_type in support_character_type, "Only {} are supported now but get {}".format(
support_character_type, self.character_str)
if character_type == "en":
self.character_str = "0123456789abcdefghijklmnopqrstuvwxyz"
dict_character = list(self.character_str)
elif character_type == "ch":
elif character_type in ["ch", "french", "german", "japan", "korean"]:
self.character_str = ""
assert character_dict_path is not None, "character_dict_path should not be None when character_type is ch"
with open(character_dict_path, "rb") as fin:
@ -150,4 +152,4 @@ class AttnLabelDecode(BaseRecLabelDecode):
else:
assert False, "unsupport type %s in get_beg_end_flag_idx" \
% beg_or_end
return idx
return idx

Loading…
Cancel
Save