Merge pull request #7960 from andyjpaddle/fix_vl

[cp]7957

doc/doc_ch/algorithm_rec_visionlan.md

@@ -139,7 +139,7 @@ Predicts of ./doc/imgs_words/en/word_2.png:('yourself', 0.9999493)
 ## 5. FAQ
 
 1. MJSynth和SynthText两种数据集来自于[VisionLAN源repo](https://github.com/wangyuxin87/VisionLAN) 。
-2. 我们使用VisionLAN作者提供的预训练模型进行finetune训练。
+2. 我们使用VisionLAN作者提供的预训练模型进行finetune训练，预训练模型配套字典为'ppocr/utils/ic15_dict.txt'。
 
 ## 引用
 

doc/doc_en/algorithm_rec_visionlan_en.md

@@ -120,7 +120,7 @@ Not supported
 ## 5. FAQ
 
 1. Note that the MJSynth and SynthText datasets come from [VisionLAN repo](https://github.com/wangyuxin87/VisionLAN).
-2. We use the pre-trained model provided by the VisionLAN authors for finetune training.
+2. We use the pre-trained model provided by the VisionLAN authors for finetune training. The dictionary for the pre-trained model is 'ppocr/utils/ic15_dict.txt'.
 
 ## Citation
 

ppocr/data/imaug/label_ops.py

@@ -107,6 +107,7 @@ class BaseRecLabelEncode(object):
         self.beg_str = "sos"
         self.end_str = "eos"
         self.lower = lower
+        self.use_default_dict = False
 
         if character_dict_path is None:
             logger = get_logger()
@@ -116,8 +117,11 @@ class BaseRecLabelEncode(object):
             self.character_str = "0123456789abcdefghijklmnopqrstuvwxyz"
             dict_character = list(self.character_str)
             self.lower = True
+            self.use_default_dict = True
         else:
             self.character_str = []
+            if 'ppocr/utils/ic15_dict.txt' in character_dict_path:
+                self.use_default_dict = True
             with open(character_dict_path, "rb") as fin:
                 lines = fin.readlines()
                 for line in lines:
@@ -1344,6 +1348,7 @@ class VLLabelEncode(BaseRecLabelEncode):
                  **kwargs):
         super(VLLabelEncode, self).__init__(
             max_text_length, character_dict_path, use_space_char, lower)
+        if self.use_default_dict:
             self.character = self.character[10:] + self.character[
                 1:10] + [self.character[0]]
         self.dict = {}

ppocr/postprocess/rec_postprocess.py

@@ -26,10 +26,15 @@ class BaseRecLabelDecode(object):
         self.end_str = "eos"
         self.reverse = False
         self.character_str = []
+        self.use_default_dict = False
+
         if character_dict_path is None:
             self.character_str = "0123456789abcdefghijklmnopqrstuvwxyz"
             dict_character = list(self.character_str)
+            self.use_default_dict = True
         else:
+            if 'ppocr/utils/ic15_dict.txt' in character_dict_path:
+                self.use_default_dict = True
             with open(character_dict_path, "rb") as fin:
                 lines = fin.readlines()
                 for line in lines:
@@ -720,6 +725,7 @@ class VLLabelDecode(BaseRecLabelDecode):
         super(VLLabelDecode, self).__init__(character_dict_path, use_space_char)
         self.max_text_length = kwargs.get('max_text_length', 25)
         self.nclass = len(self.character) + 1
+        if self.use_default_dict:
             self.character = self.character[10:] + self.character[
                 1:10] + [self.character[0]]