Merge branch 'dygraph' into dygraph_for_srn

ppocr/data/__init__.py

@@ -65,6 +65,7 @@ def build_dataloader(config, mode, device, logger, seed=None):
     loader_config = config[mode]['loader']
     batch_size = loader_config['batch_size_per_card']
     drop_last = loader_config['drop_last']
+    shuffle = loader_config['shuffle']
     num_workers = loader_config['num_workers']
     if 'use_shared_memory' in loader_config.keys():
         use_shared_memory = loader_config['use_shared_memory']
@@ -75,14 +76,14 @@ def build_dataloader(config, mode, device, logger, seed=None):
         batch_sampler = DistributedBatchSampler(
             dataset=dataset,
             batch_size=batch_size,
-            shuffle=False,
+            shuffle=shuffle,
             drop_last=drop_last)
     else:
         #Distribute data to single card
         batch_sampler = BatchSampler(
             dataset=dataset,
             batch_size=batch_size,
-            shuffle=False,
+            shuffle=shuffle,
             drop_last=drop_last)
 
     data_loader = DataLoader(

ppocr/data/imaug/label_ops.py

@@ -18,6 +18,7 @@ from __future__ import print_function
 from __future__ import unicode_literals
 
 import numpy as np
+import string
 
 
 class ClsLabelEncode(object):
@@ -92,7 +93,10 @@ class BaseRecLabelEncode(object):
                  character_type='ch',
                  use_space_char=False):
         support_character_type = [
-            'ch', 'en', 'en_sensitive', 'french', 'german', 'japan', 'korean'
+            'ch', 'en', 'EN_symbol', 'french', 'german', 'japan', 'korean',
+            'EN', 'it', 'xi', 'pu', 'ru', 'ar', 'ta', 'ug', 'fa', 'ur', 'rs',
+            'oc', 'rsc', 'bg', 'uk', 'be', 'te', 'ka', 'chinese_cht', 'hi',
+            'mr', 'ne'
         ]
         assert character_type in support_character_type, "Only {} are supported now but get {}".format(
             support_character_type, character_type)
@@ -103,9 +107,14 @@ class BaseRecLabelEncode(object):
         if character_type == "en":
             self.character_str = "0123456789abcdefghijklmnopqrstuvwxyz"
             dict_character = list(self.character_str)
-        elif character_type in ["ch", "french", "german", "japan", "korean"]:
+        elif character_type == "EN_symbol":
+            # same with ASTER setting (use 94 char).
+            self.character_str = string.printable[:-6]
+            dict_character = list(self.character_str)
+        elif character_type in support_character_type:
             self.character_str = ""
-            assert character_dict_path is not None, "character_dict_path should not be None when character_type is ch"
+            assert character_dict_path is not None, "character_dict_path should not be None when character_type is {}".format(
+                character_type)
             with open(character_dict_path, "rb") as fin:
                 lines = fin.readlines()
                 for line in lines:
@@ -114,11 +123,6 @@ class BaseRecLabelEncode(object):
             if use_space_char:
                 self.character_str += " "
             dict_character = list(self.character_str)
-        elif character_type == "en_sensitive":
-            # same with ASTER setting (use 94 char).
-            import string
-            self.character_str = string.printable[:-6]
-            dict_character = list(self.character_str)
         self.character_type = character_type
         dict_character = self.add_special_char(dict_character)
         self.dict = {}