Merge branch 'PaddlePaddle:dygraph' into dygraph

doc/doc_ch/training.md

@@ -129,3 +129,9 @@ PaddleOCR主要聚焦通用OCR，如果有垂类需求，您可以用PaddleOCR+
 
     A：识别模型训练初期acc为0是正常的，多训一段时间指标就上来了。
 
+
+***
+具体的训练教程可点击下方链接跳转：  
+- [文本检测模型训练](https://github.com/PaddlePaddle/PaddleOCR/blob/release/2.3/doc/doc_ch/detection.md)  
+- [文本识别模型训练](https://github.com/PaddlePaddle/PaddleOCR/blob/release/2.3/doc/doc_ch/recognition.md)  
+- [文本方向分类器训练](https://github.com/PaddlePaddle/PaddleOCR/blob/release/2.3/doc/doc_ch/angle_class.md)  
\ No newline at end of file

doc/doc_en/config_en.md

@@ -37,9 +37,8 @@ Take rec_chinese_lite_train_v2.0.yml as an example
 |      checkpoints         |    set model parameter path            |       None        |   Used to load parameters after interruption to continue training|
 |      use_visualdl  |    Set whether to enable visualdl for visual log display |          False        |    [Tutorial](https://www.paddlepaddle.org.cn/paddle/visualdl) |
 |      infer_img            |    Set inference image path or folder path     |       ./infer_img | \|
-|      character_dict_path |    Set dictionary path            |  ./ppocr/utils/ppocr_keys_v1.txt  |    \                 |
+|      character_dict_path |    Set dictionary path            |  ./ppocr/utils/ppocr_keys_v1.txt  |    If the character_dict_path is None, model can only recognize number and lower letters                 |
 |      max_text_length     |    Set the maximum length of text        |       25          |                \                 |
-|      character_type      |    Set character type            |       ch          |    en/ch, the default dict will be used for en, and the custom dict will be used for ch |
 |      use_space_char     |    Set whether to recognize spaces             |        True      |          Only support in character_type=ch mode                 |
 |      label_list          |    Set the angle supported by the direction classifier       |    ['0','180']    |     Only valid in angle classifier model |
 |      save_res_path          |    Set the save address of the test model results       |    ./output/det_db/predicts_db.txt    |     Only valid in the text detection model |
@@ -196,7 +195,6 @@ Italian is made up of Latin letters, so after executing the command, you will ge
       use_gpu: True
       epoch_num: 500
       ...
-      character_type: it  # language
       character_dict_path:  {path/of/dict} # path of dict
 
    Train:
@@ -218,18 +216,18 @@ Italian is made up of Latin letters, so after executing the command, you will ge
 
 Currently, the multi-language algorithms supported by PaddleOCR are:
 
-| Configuration file |  Algorithm name |   backbone |   trans   |   seq      |     pred     |  language | character_type |
-| :--------: |  :-------:   | :-------:  |   :-------:   |   :-----:   |  :-----:   | :-----:  | :-----:  |
-| rec_chinese_cht_lite_train.yml |  CRNN |   Mobilenet_v3 small 0.5 |  None   |  BiLSTM |  ctc  | chinese traditional  | chinese_cht|
-| rec_en_lite_train.yml |  CRNN |   Mobilenet_v3 small 0.5 |  None   |  BiLSTM |  ctc  | English(Case sensitive)   | EN |
-| rec_french_lite_train.yml |  CRNN |   Mobilenet_v3 small 0.5 |  None   |  BiLSTM |  ctc  | French |  french |
-| rec_ger_lite_train.yml |  CRNN |   Mobilenet_v3 small 0.5 |  None   |  BiLSTM |  ctc  | German   | german |
-| rec_japan_lite_train.yml |  CRNN |   Mobilenet_v3 small 0.5 |  None   |  BiLSTM |  ctc  | Japanese | japan |
-| rec_korean_lite_train.yml |  CRNN |   Mobilenet_v3 small 0.5 |  None   |  BiLSTM |  ctc  | Korean  | korean |
-| rec_latin_lite_train.yml |  CRNN |   Mobilenet_v3 small 0.5 |  None   |  BiLSTM |  ctc  | Latin  | latin |
-| rec_arabic_lite_train.yml |  CRNN |   Mobilenet_v3 small 0.5 |  None   |  BiLSTM |  ctc  | arabic |  ar |
-| rec_cyrillic_lite_train.yml |  CRNN |   Mobilenet_v3 small 0.5 |  None   |  BiLSTM |  ctc  | cyrillic   | cyrillic |
-| rec_devanagari_lite_train.yml |  CRNN |   Mobilenet_v3 small 0.5 |  None   |  BiLSTM |  ctc  | devanagari  | devanagari |
+| Configuration file |  Algorithm name |   backbone |   trans   |   seq      |     pred     |  language |
+| :--------: |  :-------:   | :-------:  |   :-------:   |   :-----:   |  :-----:   | :-----:  |
+| rec_chinese_cht_lite_train.yml |  CRNN |   Mobilenet_v3 small 0.5 |  None   |  BiLSTM |  ctc  | chinese traditional  |
+| rec_en_lite_train.yml |  CRNN |   Mobilenet_v3 small 0.5 |  None   |  BiLSTM |  ctc  | English(Case sensitive)   |
+| rec_french_lite_train.yml |  CRNN |   Mobilenet_v3 small 0.5 |  None   |  BiLSTM |  ctc  | French |  
+| rec_ger_lite_train.yml |  CRNN |   Mobilenet_v3 small 0.5 |  None   |  BiLSTM |  ctc  | German   |
+| rec_japan_lite_train.yml |  CRNN |   Mobilenet_v3 small 0.5 |  None   |  BiLSTM |  ctc  | Japanese |
+| rec_korean_lite_train.yml |  CRNN |   Mobilenet_v3 small 0.5 |  None   |  BiLSTM |  ctc  | Korean  |
+| rec_latin_lite_train.yml |  CRNN |   Mobilenet_v3 small 0.5 |  None   |  BiLSTM |  ctc  | Latin  |
+| rec_arabic_lite_train.yml |  CRNN |   Mobilenet_v3 small 0.5 |  None   |  BiLSTM |  ctc  | arabic |  
+| rec_cyrillic_lite_train.yml |  CRNN |   Mobilenet_v3 small 0.5 |  None   |  BiLSTM |  ctc  | cyrillic   |
+| rec_devanagari_lite_train.yml |  CRNN |   Mobilenet_v3 small 0.5 |  None   |  BiLSTM |  ctc  | devanagari  |
 
 For more supported languages, please refer to : [Multi-language model](https://github.com/PaddlePaddle/PaddleOCR/blob/release/2.1/doc/doc_en/multi_languages_en.md#4-support-languages-and-abbreviations)
 

doc/doc_en/inference_en.md

@@ -281,7 +281,7 @@ python3 tools/export_model.py -c configs/det/rec_r34_vd_none_bilstm_ctc.yml -o G
 For CRNN text recognition model inference, execute the following commands:
 
 ```
-python3 tools/infer/predict_rec.py --image_dir="./doc/imgs_words_en/word_336.png" --rec_model_dir="./inference/starnet/" --rec_image_shape="3, 32, 100" --rec_char_type="en"
+python3 tools/infer/predict_rec.py --image_dir="./doc/imgs_words_en/word_336.png" --rec_model_dir="./inference/starnet/" --rec_image_shape="3, 32, 100" --rec_char_dict_path="./ppocr/utils/ic15_dict.txt"
 ```
 
 ![](../imgs_words_en/word_336.png)
@@ -314,7 +314,7 @@ with the training, such as: --rec_image_shape="1, 64, 256"
 python3 tools/infer/predict_rec.py --image_dir="./doc/imgs_words_en/word_336.png" \
                                     --rec_model_dir="./inference/srn/" \
                                     --rec_image_shape="1, 64, 256" \
-                                    --rec_char_type="en" \
+                                    --rec_char_dict_path="./ppocr/utils/ic15_dict.txt" \
                                     --rec_algorithm="SRN"
 ```
 
@@ -323,7 +323,7 @@ python3 tools/infer/predict_rec.py --image_dir="./doc/imgs_words_en/word_336.png
 If the text dictionary is modified during training, when using the inference model to predict, you need to specify the dictionary path used by `--rec_char_dict_path`, and set `rec_char_type=ch`
 
 ```
-python3 tools/infer/predict_rec.py --image_dir="./doc/imgs_words_en/word_336.png" --rec_model_dir="./your inference model" --rec_image_shape="3, 32, 100" --rec_char_type="ch" --rec_char_dict_path="your text dict path"
+python3 tools/infer/predict_rec.py --image_dir="./doc/imgs_words_en/word_336.png" --rec_model_dir="./your inference model" --rec_image_shape="3, 32, 100"  --rec_char_dict_path="your text dict path"
 ```
 
 <a name="MULTILINGUAL_MODEL_INFERENCE"></a>
@@ -333,7 +333,7 @@ If you need to predict other language models, when using inference model predict
 You need to specify the visual font path through `--vis_font_path`. There are small language fonts provided by default under the `doc/fonts` path, such as Korean recognition:
 
 ```
-python3 tools/infer/predict_rec.py --image_dir="./doc/imgs_words/korean/1.jpg" --rec_model_dir="./your inference model" --rec_char_type="korean" --rec_char_dict_path="ppocr/utils/dict/korean_dict.txt" --vis_font_path="doc/fonts/korean.ttf"
+python3 tools/infer/predict_rec.py --image_dir="./doc/imgs_words/korean/1.jpg" --rec_model_dir="./your inference model"  --rec_char_dict_path="ppocr/utils/dict/korean_dict.txt" --vis_font_path="doc/fonts/korean.ttf"
 ```
 ![](../imgs_words/korean/1.jpg)
 
@@ -399,7 +399,7 @@ If you want to try other detection algorithms or recognition algorithms, please
 The following command uses the combination of the EAST text detection and STAR-Net text recognition:
 
 ```
-python3 tools/infer/predict_system.py --image_dir="./doc/imgs_en/img_10.jpg" --det_model_dir="./inference/det_east/" --det_algorithm="EAST" --rec_model_dir="./inference/starnet/" --rec_image_shape="3, 32, 100" --rec_char_type="en"
+python3 tools/infer/predict_system.py --image_dir="./doc/imgs_en/img_10.jpg" --det_model_dir="./inference/det_east/" --det_algorithm="EAST" --rec_model_dir="./inference/starnet/" --rec_image_shape="3, 32, 100" --rec_char_dict_path="./ppocr/utils/ic15_dict.txt"
 ```
 
 After executing the command, the recognition result image is as follows:

doc/doc_en/recognition_en.md

@@ -161,7 +161,7 @@ The current multi-language model is still in the demo stage and will continue to
 If you like, you can submit the dictionary file to [dict](../../ppocr/utils/dict) and we will thank you in the Repo.
 
 
-To customize the dict file, please modify the `character_dict_path` field in `configs/rec/rec_icdar15_train.yml` and set `character_type` to `ch`.
+To customize the dict file, please modify the `character_dict_path` field in `configs/rec/rec_icdar15_train.yml` .
 
 - Custom dictionary
 
@@ -172,8 +172,6 @@ If you need to customize dic file, please add character_dict_path field in confi
 
 If you want to support the recognition of the `space` category, please set the `use_space_char` field in the yml file to `True`.
 
-**Note: use_space_char only takes effect when character_type=ch**
-
 <a name="TRAINING"></a>
 ## 2.Training
 
@@ -250,7 +248,6 @@ Global:
   # Add a custom dictionary, such as modify the dictionary, please point the path to the new dictionary
   character_dict_path: ppocr/utils/ppocr_keys_v1.txt
   # Modify character type
-  character_type: ch
   ...
   # Whether to recognize spaces
   use_space_char: True
@@ -312,18 +309,18 @@ Eval:
 
 Currently, the multi-language algorithms supported by PaddleOCR are:
 
-| Configuration file |  Algorithm name |   backbone |   trans   |   seq      |     pred     |  language | character_type |
-| :--------: |  :-------:   | :-------:  |   :-------:   |   :-----:   |  :-----:   | :-----:  | :-----:  |
-| rec_chinese_cht_lite_train.yml |  CRNN |   Mobilenet_v3 small 0.5 |  None   |  BiLSTM |  ctc  | chinese traditional  | chinese_cht|
-| rec_en_lite_train.yml |  CRNN |   Mobilenet_v3 small 0.5 |  None   |  BiLSTM |  ctc  | English(Case sensitive)   | EN |
-| rec_french_lite_train.yml |  CRNN |   Mobilenet_v3 small 0.5 |  None   |  BiLSTM |  ctc  | French |  french |
-| rec_ger_lite_train.yml |  CRNN |   Mobilenet_v3 small 0.5 |  None   |  BiLSTM |  ctc  | German   | german |
-| rec_japan_lite_train.yml |  CRNN |   Mobilenet_v3 small 0.5 |  None   |  BiLSTM |  ctc  | Japanese | japan |
-| rec_korean_lite_train.yml |  CRNN |   Mobilenet_v3 small 0.5 |  None   |  BiLSTM |  ctc  | Korean  | korean |
-| rec_latin_lite_train.yml |  CRNN |   Mobilenet_v3 small 0.5 |  None   |  BiLSTM |  ctc  | Latin  | latin |
-| rec_arabic_lite_train.yml |  CRNN |   Mobilenet_v3 small 0.5 |  None   |  BiLSTM |  ctc  | arabic |  ar |
-| rec_cyrillic_lite_train.yml |  CRNN |   Mobilenet_v3 small 0.5 |  None   |  BiLSTM |  ctc  | cyrillic   | cyrillic |
-| rec_devanagari_lite_train.yml |  CRNN |   Mobilenet_v3 small 0.5 |  None   |  BiLSTM |  ctc  | devanagari  | devanagari |
+| Configuration file |  Algorithm name |   backbone |   trans   |   seq      |     pred     |  language |
+| :--------: |  :-------:   | :-------:  |   :-------:   |   :-----:   |  :-----:   | :-----:  |
+| rec_chinese_cht_lite_train.yml |  CRNN |   Mobilenet_v3 small 0.5 |  None   |  BiLSTM |  ctc  | chinese traditional  |
+| rec_en_lite_train.yml |  CRNN |   Mobilenet_v3 small 0.5 |  None   |  BiLSTM |  ctc  | English(Case sensitive)   |
+| rec_french_lite_train.yml |  CRNN |   Mobilenet_v3 small 0.5 |  None   |  BiLSTM |  ctc  | French |  
+| rec_ger_lite_train.yml |  CRNN |   Mobilenet_v3 small 0.5 |  None   |  BiLSTM |  ctc  | German   |
+| rec_japan_lite_train.yml |  CRNN |   Mobilenet_v3 small 0.5 |  None   |  BiLSTM |  ctc  | Japanese |
+| rec_korean_lite_train.yml |  CRNN |   Mobilenet_v3 small 0.5 |  None   |  BiLSTM |  ctc  | Korean  |
+| rec_latin_lite_train.yml |  CRNN |   Mobilenet_v3 small 0.5 |  None   |  BiLSTM |  ctc  | Latin  |
+| rec_arabic_lite_train.yml |  CRNN |   Mobilenet_v3 small 0.5 |  None   |  BiLSTM |  ctc  | arabic |
+| rec_cyrillic_lite_train.yml |  CRNN |   Mobilenet_v3 small 0.5 |  None   |  BiLSTM |  ctc  | cyrillic   |
+| rec_devanagari_lite_train.yml |  CRNN |   Mobilenet_v3 small 0.5 |  None   |  BiLSTM |  ctc  | devanagari  |
 
 For more supported languages, please refer to : [Multi-language model](https://github.com/PaddlePaddle/PaddleOCR/blob/release/2.1/doc/doc_en/multi_languages_en.md#4-support-languages-and-abbreviations)
 
@@ -471,6 +468,3 @@ inference/det_db/
   ```
   python3 tools/infer/predict_rec.py --image_dir="./doc/imgs_words_en/word_336.png" --rec_model_dir="./your inference model" --rec_image_shape="3, 32, 100" --rec_char_type="ch" --rec_char_dict_path="your text dict path"
   ```
-
-  
-

doc/doc_en/training_en.md

@@ -147,3 +147,9 @@ There are several experiences for reference when constructing the data set:
 
     A: It is normal for the acc to be 0 at the beginning of the recognition model training, and the indicator will come up after a longer training period.
 
+
+***
+Click the following links for detailed training tutorial:  
+- [text detection model training](https://github.com/PaddlePaddle/PaddleOCR/blob/release/2.3/doc/doc_ch/detection.md)  
+- [text recognition model training](https://github.com/PaddlePaddle/PaddleOCR/blob/release/2.3/doc/doc_ch/recognition.md)  
+- [text direction classification model training](https://github.com/PaddlePaddle/PaddleOCR/blob/release/2.3/doc/doc_ch/angle_class.md)  

ppocr/data/imaug/label_ops.py

@@ -21,6 +21,8 @@ import numpy as np
 import string
 import json
 
+from ppocr.utils.logging import get_logger
+
 
 class ClsLabelEncode(object):
     def __init__(self, label_list, **kwargs):
@@ -92,31 +94,23 @@ class BaseRecLabelEncode(object):
     def __init__(self,
                  max_text_length,
                  character_dict_path=None,
-                 character_type='ch',
                  use_space_char=False):
-        support_character_type = [
-            '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', 'latin', 'arabic', 'cyrillic', 'devanagari'
-        ]
-        assert character_type in support_character_type, "Only {} are supported now but get {}".format(
-            support_character_type, character_type)
 
         self.max_text_len = max_text_length
         self.beg_str = "sos"
         self.end_str = "eos"
-        if character_type == "en":
+        self.lower = False
+
+        if character_dict_path is None:
+            logger = get_logger()
+            logger.warning(
+                "The character_dict_path is None, model can only recognize number and lower letters"
+            )
             self.character_str = "0123456789abcdefghijklmnopqrstuvwxyz"
             dict_character = list(self.character_str)
-        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.lower = True
+        else:
             self.character_str = ""
-            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:
@@ -125,7 +119,6 @@ class BaseRecLabelEncode(object):
             if use_space_char:
                 self.character_str += " "
             dict_character = list(self.character_str)
-        self.character_type = character_type
         dict_character = self.add_special_char(dict_character)
         self.dict = {}
         for i, char in enumerate(dict_character):
@@ -147,7 +140,7 @@ class BaseRecLabelEncode(object):
         """
         if len(text) == 0 or len(text) > self.max_text_len:
             return None
-        if self.character_type == "en":
+        if self.lower:
             text = text.lower()
         text_list = []
         for char in text:
@@ -167,13 +160,11 @@ class NRTRLabelEncode(BaseRecLabelEncode):
     def __init__(self,
                  max_text_length,
                  character_dict_path=None,
-                 character_type='EN_symbol',
                  use_space_char=False,
                  **kwargs):
 
-        super(NRTRLabelEncode,
-              self).__init__(max_text_length, character_dict_path,
-                             character_type, use_space_char)
+        super(NRTRLabelEncode, self).__init__(
+            max_text_length, character_dict_path, use_space_char)
 
     def __call__(self, data):
         text = data['label']
@@ -200,12 +191,10 @@ class CTCLabelEncode(BaseRecLabelEncode):
     def __init__(self,
                  max_text_length,
                  character_dict_path=None,
-                 character_type='ch',
                  use_space_char=False,
                  **kwargs):
-        super(CTCLabelEncode,
-              self).__init__(max_text_length, character_dict_path,
-                             character_type, use_space_char)
+        super(CTCLabelEncode, self).__init__(
+            max_text_length, character_dict_path, use_space_char)
 
     def __call__(self, data):
         text = data['label']
@@ -231,12 +220,10 @@ class E2ELabelEncodeTest(BaseRecLabelEncode):
     def __init__(self,
                  max_text_length,
                  character_dict_path=None,
-                 character_type='EN',
                  use_space_char=False,
                  **kwargs):
-        super(E2ELabelEncodeTest,
-              self).__init__(max_text_length, character_dict_path,
-                             character_type, use_space_char)
+        super(E2ELabelEncodeTest, self).__init__(
+            max_text_length, character_dict_path, use_space_char)
 
     def __call__(self, data):
         import json
@@ -305,12 +292,10 @@ class AttnLabelEncode(BaseRecLabelEncode):
     def __init__(self,
                  max_text_length,
                  character_dict_path=None,
-                 character_type='ch',
                  use_space_char=False,
                  **kwargs):
-        super(AttnLabelEncode,
-              self).__init__(max_text_length, character_dict_path,
-                             character_type, use_space_char)
+        super(AttnLabelEncode, self).__init__(
+            max_text_length, character_dict_path, use_space_char)
 
     def add_special_char(self, dict_character):
         self.beg_str = "sos"
@@ -353,12 +338,10 @@ class SEEDLabelEncode(BaseRecLabelEncode):
     def __init__(self,
                  max_text_length,
                  character_dict_path=None,
-                 character_type='ch',
                  use_space_char=False,
                  **kwargs):
-        super(SEEDLabelEncode,
-              self).__init__(max_text_length, character_dict_path,
-                             character_type, use_space_char)
+        super(SEEDLabelEncode, self).__init__(
+            max_text_length, character_dict_path, use_space_char)
 
     def add_special_char(self, dict_character):
         self.end_str = "eos"
@@ -385,12 +368,10 @@ class SRNLabelEncode(BaseRecLabelEncode):
     def __init__(self,
                  max_text_length=25,
                  character_dict_path=None,
-                 character_type='en',
                  use_space_char=False,
                  **kwargs):
-        super(SRNLabelEncode,
-              self).__init__(max_text_length, character_dict_path,
-                             character_type, use_space_char)
+        super(SRNLabelEncode, self).__init__(
+            max_text_length, character_dict_path, use_space_char)
 
     def add_special_char(self, dict_character):
         dict_character = dict_character + [self.beg_str, self.end_str]
@@ -598,12 +579,10 @@ class SARLabelEncode(BaseRecLabelEncode):
     def __init__(self,
                  max_text_length,
                  character_dict_path=None,
-                 character_type='ch',
                  use_space_char=False,
                  **kwargs):
-        super(SARLabelEncode,
-              self).__init__(max_text_length, character_dict_path,
-                             character_type, use_space_char)
+        super(SARLabelEncode, self).__init__(
+            max_text_length, character_dict_path, use_space_char)
 
     def add_special_char(self, dict_character):
         beg_end_str = "<BOS/EOS>"

ppocr/data/imaug/rec_img_aug.py

@@ -87,17 +87,17 @@ class RecResizeImg(object):
     def __init__(self,
                  image_shape,
                  infer_mode=False,
-                 character_type='ch',
+                 character_dict_path='./ppocr/utils/ppocr_keys_v1.txt',
                  padding=True,
                  **kwargs):
         self.image_shape = image_shape
         self.infer_mode = infer_mode
-        self.character_type = character_type
+        self.character_dict_path = character_dict_path
         self.padding = padding
 
     def __call__(self, data):
         img = data['image']
-        if self.infer_mode and self.character_type == "ch":
+        if self.infer_mode and self.character_dict_path is not None:
             norm_img = resize_norm_img_chinese(img, self.image_shape)
         else:
             norm_img = resize_norm_img(img, self.image_shape, self.padding)

ppocr/postprocess/rec_postprocess.py

@@ -21,33 +21,15 @@ import re
 class BaseRecLabelDecode(object):
     """ Convert between text-label and text-index """
 
-    def __init__(self,
-                 character_dict_path=None,
-                 character_type='ch',
-                 use_space_char=False):
-        support_character_type = [
-            'ch', 'en', 'EN_symbol', 'french', 'german', 'japan', 'korean',
-            'it', 'xi', 'pu', 'ru', 'ar', 'ta', 'ug', 'fa', 'ur', 'rs', 'oc',
-            'rsc', 'bg', 'uk', 'be', 'te', 'ka', 'chinese_cht', 'hi', 'mr',
-            'ne', 'EN', 'latin', 'arabic', 'cyrillic', 'devanagari'
-        ]
-        assert character_type in support_character_type, "Only {} are supported now but get {}".format(
-            support_character_type, character_type)
-
+    def __init__(self, character_dict_path=None, use_space_char=False):
         self.beg_str = "sos"
         self.end_str = "eos"
 
-        if character_type == "en":
+        self.character_str = []
+        if character_dict_path is None:
             self.character_str = "0123456789abcdefghijklmnopqrstuvwxyz"
             dict_character = list(self.character_str)
-        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 {}".format(
-                character_type)
+        else:
             with open(character_dict_path, "rb") as fin:
                 lines = fin.readlines()
                 for line in lines:
@@ -57,9 +39,6 @@ class BaseRecLabelDecode(object):
                 self.character_str.append(" ")
             dict_character = list(self.character_str)
 
-        else:
-            raise NotImplementedError
-        self.character_type = character_type
         dict_character = self.add_special_char(dict_character)
         self.dict = {}
         for i, char in enumerate(dict_character):
@@ -102,13 +81,10 @@ class BaseRecLabelDecode(object):
 class CTCLabelDecode(BaseRecLabelDecode):
     """ Convert between text-label and text-index """
 
-    def __init__(self,
-                 character_dict_path=None,
-                 character_type='ch',
-                 use_space_char=False,
+    def __init__(self, character_dict_path=None, use_space_char=False,
                  **kwargs):
         super(CTCLabelDecode, self).__init__(character_dict_path,
-                                             character_type, use_space_char)
+                                             use_space_char)
 
     def __call__(self, preds, label=None, *args, **kwargs):
         if isinstance(preds, tuple):
@@ -136,13 +112,12 @@ class DistillationCTCLabelDecode(CTCLabelDecode):
 
     def __init__(self,
                  character_dict_path=None,
-                 character_type='ch',
                  use_space_char=False,
                  model_name=["student"],
                  key=None,
                  **kwargs):
-        super(DistillationCTCLabelDecode, self).__init__(
-            character_dict_path, character_type, use_space_char)
+        super(DistillationCTCLabelDecode, self).__init__(character_dict_path,
+                                                         use_space_char)
         if not isinstance(model_name, list):
             model_name = [model_name]
         self.model_name = model_name
@@ -162,13 +137,9 @@ class DistillationCTCLabelDecode(CTCLabelDecode):
 class NRTRLabelDecode(BaseRecLabelDecode):
     """ Convert between text-label and text-index """
 
-    def __init__(self,
-                 character_dict_path=None,
-                 character_type='EN_symbol',
-                 use_space_char=True,
-                 **kwargs):
+    def __init__(self, character_dict_path=None, use_space_char=True, **kwargs):
         super(NRTRLabelDecode, self).__init__(character_dict_path,
-                                              character_type, use_space_char)
+                                              use_space_char)
 
     def __call__(self, preds, label=None, *args, **kwargs):
 
@@ -230,13 +201,10 @@ class NRTRLabelDecode(BaseRecLabelDecode):
 class AttnLabelDecode(BaseRecLabelDecode):
     """ Convert between text-label and text-index """
 
-    def __init__(self,
-                 character_dict_path=None,
-                 character_type='ch',
-                 use_space_char=False,
+    def __init__(self, character_dict_path=None, use_space_char=False,
                  **kwargs):
         super(AttnLabelDecode, self).__init__(character_dict_path,
-                                              character_type, use_space_char)
+                                              use_space_char)
 
     def add_special_char(self, dict_character):
         self.beg_str = "sos"
@@ -313,13 +281,10 @@ class AttnLabelDecode(BaseRecLabelDecode):
 class SEEDLabelDecode(BaseRecLabelDecode):
     """ Convert between text-label and text-index """
 
-    def __init__(self,
-                 character_dict_path=None,
-                 character_type='ch',
-                 use_space_char=False,
+    def __init__(self, character_dict_path=None, use_space_char=False,
                  **kwargs):
         super(SEEDLabelDecode, self).__init__(character_dict_path,
-                                              character_type, use_space_char)
+                                              use_space_char)
 
     def add_special_char(self, dict_character):
         self.beg_str = "sos"
@@ -394,13 +359,10 @@ class SEEDLabelDecode(BaseRecLabelDecode):
 class SRNLabelDecode(BaseRecLabelDecode):
     """ Convert between text-label and text-index """
 
-    def __init__(self,
-                 character_dict_path=None,
-                 character_type='en',
-                 use_space_char=False,
+    def __init__(self, character_dict_path=None, use_space_char=False,
                  **kwargs):
         super(SRNLabelDecode, self).__init__(character_dict_path,
-                                             character_type, use_space_char)
+                                             use_space_char)
         self.max_text_length = kwargs.get('max_text_length', 25)
 
     def __call__(self, preds, label=None, *args, **kwargs):
@@ -616,13 +578,10 @@ class TableLabelDecode(object):
 class SARLabelDecode(BaseRecLabelDecode):
     """ Convert between text-label and text-index """
 
-    def __init__(self,
-                 character_dict_path=None,
-                 character_type='ch',
-                 use_space_char=False,
+    def __init__(self, character_dict_path=None, use_space_char=False,
                  **kwargs):
         super(SARLabelDecode, self).__init__(character_dict_path,
-                                             character_type, use_space_char)
+                                             use_space_char)
 
         self.rm_symbol = kwargs.get('rm_symbol', False)
 

ppocr/utils/EN_symbol_dict.txt

+0
+1
+2
+3
+4
+5
+6
+7
+8
+9
+a
+b
+c
+d
+e
+f
+g
+h
+i
+j
+k
+l
+m
+n
+o
+p
+q
+r
+s
+t
+u
+v
+w
+x
+y
+z
+A
+B
+C
+D
+E
+F
+G
+H
+I
+J
+K
+L
+M
+N
+O
+P
+Q
+R
+S
+T
+U
+V
+W
+X
+Y
+Z
+!
+"
+#
+$
+%
+&
+'
+(
+)
+*
++
+,
+-
+.
+/
+:
+;
+<
+=
+>
+?
+@
+[
+\
+]
+^
+_
+`
+{
+|
+}
+~
\ No newline at end of file

requirements.txt

 shapely
-scikit-image==0.17.2
+scikit-image==0.18.3
 imgaug==0.4.0
 pyclipper
 lmdb

tests/docs/test_cpp.md

-# C++预测功能测试
-
-C++预测功能测试的主程序为`test_cpp.sh`，可以测试基于C++预测库的模型推理功能。
-
-## 测试结论汇总
-
-| 算法名称 | 模型名称 |device | batchsize | mkldnn | cpu多线程 | tensorrt | 离线量化 |
-|  ----  |   ----  |  ----  |  ---- |  ---- |  ----  |  ----| --- | 
-| DB   |ch_ppocr_mobile_v2.0_det| CPU/GPU | 1/6 | 支持 | 支持 | fp32/fp16/int8 | 支持 |
-| DB   |ch_ppocr_server_v2.0_det| CPU/GPU | 1/6 | 支持 | 支持 | fp32/fp16/int8 | 支持 |
-| CRNN |ch_ppocr_mobile_v2.0_rec| CPU/GPU | 1/6 | 支持 | 支持 | fp32/fp16/int8 | 支持 |
-| CRNN |ch_ppocr_server_v2.0_rec| CPU/GPU | 1/6 | 支持 | 支持 | fp32/fp16/int8 | 支持 |
-|PP-OCR|ch_ppocr_server_v2.0    | CPU/GPU | 1/6 | 支持 | 支持 | fp32/fp16/int8 | 支持 |
-|PP-OCR|ch_ppocr_server_v2.0    | CPU/GPU | 1/6 | 支持 | 支持 | fp32/fp16/int8 | 支持 |
-
-
-
-## 1. 功能测试
-先运行`prepare.sh`准备数据和模型，然后运行`test_cpp.sh`进行测试，最终在```tests/output```目录下生成`cpp_infer_*.log`后缀的日志文件。
-
-```shell
-bash tests/prepare.sh ./tests/configs/ppocr_det_mobile_params.txt
-
-# 用法1:
-bash tests/test_cpp.sh ./tests/configs/ppocr_det_mobile_params.txt
-# 用法2: 指定GPU卡预测，第三个传入参数为GPU卡号
-bash tests/test_cpp.sh ./tests/configs/ppocr_det_mobile_params.txt '1'
-```  
- 
-
-## 2. 精度测试
-
-使用compare_results.py脚本比较模型预测的结果是否符合预期，主要步骤包括：
-- 提取日志中的预测坐标；
-- 从本地文件中提取保存好的坐标结果；
-- 比较上述两个结果是否符合精度预期，误差大于设置阈值时会报错。
-
-### 使用方式
-运行命令：
-```shell
-python3.7 tests/compare_results.py --gt_file=./tests/results/*.txt  --log_file=./tests/output/infer_*.log --atol=1e-3 --rtol=1e-3
-```
-
-参数介绍：  
-- gt_file： 指向事先保存好的预测结果路径，支持*.txt 结尾，会自动索引*.txt格式的文件，文件默认保存在tests/result/ 文件夹下
-- log_file: 指向运行tests/test.sh 脚本的infer模式保存的预测日志，预测日志中打印的有预测结果，比如：文本框，预测文本，类别等等，同样支持infer_*.log格式传入
-- atol: 设置的绝对误差
-- rtol: 设置的相对误差
-
-### 运行结果
-
-正常运行效果如下图：
-<img src="compare_right.png" width="1000">
-
-出现不一致结果时的运行输出：
-<img src="compare_wrong.png" width="1000">

tools/infer/predict_cls.py

@@ -131,14 +131,9 @@ def main(args):
         img_list.append(img)
     try:
         img_list, cls_res, predict_time = text_classifier(img_list)
-    except:
+    except Exception as E:
         logger.info(traceback.format_exc())
-        logger.info(
-            "ERROR!!!! \n"
-            "Please read the FAQ：https://github.com/PaddlePaddle/PaddleOCR#faq \n"
-            "If your model has tps module:  "
-            "TPS does not support variable shape.\n"
-            "Please set --rec_image_shape='3,32,100' and --rec_char_type='en' ")
+        logger.info(E)
         exit()
     for ino in range(len(img_list)):
         logger.info("Predicts of {}:{}".format(valid_image_file_list[ino],

tools/infer/predict_rec.py

@@ -38,40 +38,34 @@ logger = get_logger()
 class TextRecognizer(object):
     def __init__(self, args):
         self.rec_image_shape = [int(v) for v in args.rec_image_shape.split(",")]
-        self.character_type = args.rec_char_type
         self.rec_batch_num = args.rec_batch_num
         self.rec_algorithm = args.rec_algorithm
         postprocess_params = {
             'name': 'CTCLabelDecode',
-            "character_type": args.rec_char_type,
             "character_dict_path": args.rec_char_dict_path,
             "use_space_char": args.use_space_char
         }
         if self.rec_algorithm == "SRN":
             postprocess_params = {
                 'name': 'SRNLabelDecode',
-                "character_type": args.rec_char_type,
                 "character_dict_path": args.rec_char_dict_path,
                 "use_space_char": args.use_space_char
             }
         elif self.rec_algorithm == "RARE":
             postprocess_params = {
                 'name': 'AttnLabelDecode',
-                "character_type": args.rec_char_type,
                 "character_dict_path": args.rec_char_dict_path,
                 "use_space_char": args.use_space_char
             }
         elif self.rec_algorithm == 'NRTR':
             postprocess_params = {
                 'name': 'NRTRLabelDecode',
-                "character_type": args.rec_char_type,
                 "character_dict_path": args.rec_char_dict_path,
                 "use_space_char": args.use_space_char
             }
         elif self.rec_algorithm == "SAR":
             postprocess_params = {
                 'name': 'SARLabelDecode',
-                "character_type": args.rec_char_type,
                 "character_dict_path": args.rec_char_dict_path,
                 "use_space_char": args.use_space_char
             }

tools/infer/utility.py

@@ -74,7 +74,6 @@ def init_args():
     parser.add_argument("--rec_algorithm", type=str, default='CRNN')
     parser.add_argument("--rec_model_dir", type=str)
     parser.add_argument("--rec_image_shape", type=str, default="3, 32, 320")
-    parser.add_argument("--rec_char_type", type=str, default='ch')
     parser.add_argument("--rec_batch_num", type=int, default=6)
     parser.add_argument("--max_text_length", type=int, default=25)
     parser.add_argument(
@@ -268,10 +267,11 @@ def create_predictor(args, mode, logger):
             # cache 10 different shapes for mkldnn to avoid memory leak
             config.set_mkldnn_cache_capacity(10)
             config.enable_mkldnn()
-
+            if args.precision == "fp16":
+                config.enable_mkldnn_bfloat16()
     # enable memory optim
     config.enable_memory_optim()
-    #config.disable_glog_info()
+    config.disable_glog_info()
 
     config.delete_pass("conv_transpose_eltwiseadd_bn_fuse_pass")
     if mode == 'table':

tools/program.py

@@ -159,7 +159,8 @@ def train(config,
           eval_class,
           pre_best_model_dict,
           logger,
-          vdl_writer=None):
+          vdl_writer=None,
+          scaler=None):
     cal_metric_during_train = config['Global'].get('cal_metric_during_train',
                                                    False)
     log_smooth_window = config['Global']['log_smooth_window']
@@ -226,12 +227,27 @@ def train(config,
             images = batch[0]
             if use_srn:
                 model_average = True
+
+            # use amp
+            if scaler:
+                with paddle.amp.auto_cast():
+                    if model_type == 'table' or extra_input:
+                        preds = model(images, data=batch[1:])
+                    else:
+                        preds = model(images)
+            else:
                 if model_type == 'table' or extra_input:
                     preds = model(images, data=batch[1:])
                 else:
                     preds = model(images)
             loss = loss_class(preds, batch)
             avg_loss = loss['loss']
+
+            if scaler:
+                scaled_avg_loss = scaler.scale(avg_loss)
+                scaled_avg_loss.backward()
+                scaler.minimize(optimizer, scaled_avg_loss)
+            else:
                 avg_loss.backward()
                 optimizer.step()
             optimizer.clear_grad()

tools/train.py

@@ -102,10 +102,27 @@ def main(config, device, logger, vdl_writer):
     if valid_dataloader is not None:
         logger.info('valid dataloader has {} iters'.format(
             len(valid_dataloader)))
+
+    use_amp = config["Global"].get("use_amp", False)
+    if use_amp:
+        AMP_RELATED_FLAGS_SETTING = {
+            'FLAGS_cudnn_batchnorm_spatial_persistent': 1,
+            'FLAGS_max_inplace_grad_add': 8,
+        }
+        paddle.fluid.set_flags(AMP_RELATED_FLAGS_SETTING)
+        scale_loss = config["Global"].get("scale_loss", 1.0)
+        use_dynamic_loss_scaling = config["Global"].get(
+            "use_dynamic_loss_scaling", False)
+        scaler = paddle.amp.GradScaler(
+            init_loss_scaling=scale_loss,
+            use_dynamic_loss_scaling=use_dynamic_loss_scaling)
+    else:
+        scaler = None
+
     # start train
     program.train(config, train_dataloader, valid_dataloader, device, model,
                   loss_class, optimizer, lr_scheduler, post_process_class,
-                  eval_class, pre_best_model_dict, logger, vdl_writer)
+                  eval_class, pre_best_model_dict, logger, vdl_writer, scaler)
 
 
 def test_reader(config, device, logger):