Merge remote-tracking branch 'PaddlePaddle/dygraph' into dygraph

doc/doc_en/models_list_en.md

@@ -27,7 +27,7 @@ Relationship of the above models is as follows.
 
 |model name|description|config|model size|download|
 | --- | --- | --- | --- | --- |
-|ch_ppocr_mobile_slim_v2.0_det|Slim pruned lightweight model, supporting Chinese, English, multilingual text detection|[ch_det_mv3_db_v2.0.yml](../../configs/det/ch_ppocr_v2.0/ch_det_mv3_db_v2.0.yml)| |inference model (coming soon) / slim model (coming soon)|
+|ch_ppocr_mobile_slim_v2.0_det|Slim pruned lightweight model, supporting Chinese, English, multilingual text detection|[ch_det_mv3_db_v2.0.yml](../../configs/det/ch_ppocr_v2.0/ch_det_mv3_db_v2.0.yml)| 2.6M |[inference model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/slim/ch_ppocr_mobile_v2.0_det_prune_infer.tar)|
 |ch_ppocr_mobile_v2.0_det|Original lightweight model, supporting Chinese, English, multilingual text detection|[ch_det_mv3_db_v2.0.yml](../../configs/det/ch_ppocr_v2.0/ch_det_mv3_db_v2.0.yml)|3M|[inference model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_det_infer.tar) / [trained model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_det_train.tar)|
 |ch_ppocr_server_v2.0_det|General model, which is larger than the lightweight model, but achieved better performance|[ch_det_res18_db_v2.0.yml](../../configs/det/ch_ppocr_v2.0/ch_det_res18_db_v2.0.yml)|47M|[inference model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_server_v2.0_det_infer.tar) / [trained model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_server_v2.0_det_train.tar)|
 

doc/doc_en/multi_languages_en.md

@@ -83,32 +83,37 @@ Paddleocr currently supports 80 languages, which can be switched by modifying th
 The specific supported [language] (#language_abbreviations) can be viewed in the table.
 
 ``` bash
-
-paddleocr --image_dir doc/imgs/japan_2.jpg --lang=japan
+paddleocr --image_dir doc/imgs_en/254.jpg --lang=en
 ```
-![](https://raw.githubusercontent.com/PaddlePaddle/PaddleOCR/release/2.0/doc/imgs/japan_2.jpg)
+<div align="center">
+    <img src="../imgs_en/254.jpg" width="300" height="600">
+    <img src="../imgs_results/multi_lang/img_02.jpg" width="600" height="600">
+</div>
 
 The result is a list, each item contains a text box, text and recognition confidence
 ```text
-[[[671.0, 60.0], [847.0, 63.0], [847.0, 104.0], [671.0, 102.0]], ('もちもち', 0.9993342)]
-[[[394.0, 82.0], [536.0, 77.0], [538.0, 127.0], [396.0, 132.0]], ('自然の', 0.9919842)]
-[[[880.0, 89.0], [1014.0, 93.0], [1013.0, 127.0], [879.0, 124.0]], ('とろっと', 0.9976762)]
-[[[1067.0, 101.0], [1294.0, 101.0], [1294.0, 138.0], [1067.0, 138.0]], ('后味のよい', 0.9988712)]
+[('PHO CAPITAL', 0.95723116), [[66.0, 50.0], [327.0, 44.0], [327.0, 76.0], [67.0, 82.0]]]
+[('107 State Street', 0.96311164), [[72.0, 90.0], [451.0, 84.0], [452.0, 116.0], [73.0, 121.0]]]
+[('Montpelier Vermont', 0.97389287), [[69.0, 132.0], [501.0, 126.0], [501.0, 158.0], [70.0, 164.0]]]
+[('8022256183', 0.99810505), [[71.0, 175.0], [363.0, 170.0], [364.0, 202.0], [72.0, 207.0]]]
+[('REG 07-24-201706:59 PM', 0.93537045), [[73.0, 299.0], [653.0, 281.0], [654.0, 318.0], [74.0, 336.0]]]
+[('045555', 0.99346405), [[509.0, 331.0], [651.0, 325.0], [652.0, 356.0], [511.0, 362.0]]]
+[('CT1', 0.9988654), [[535.0, 367.0], [654.0, 367.0], [654.0, 406.0], [535.0, 406.0]]]
 ......
 ```
 
 * Recognition
 
 ```bash
-paddleocr --image_dir doc/imgs_words/japan/1.jpg --det false --lang=japan
+paddleocr --image_dir doc/imgs_words_en/word_308.png --det false --lang=en
 ```
 
-![](https://raw.githubusercontent.com/PaddlePaddle/PaddleOCR/release/2.0/doc/imgs_words/japan/1.jpg)
+![](https://raw.githubusercontent.com/PaddlePaddle/PaddleOCR/release/2.1/doc/imgs_words_en/word_308.png)
 
 The result is a tuple, which returns the recognition result and recognition confidence
 
 ```text
-('したがって', 0.99965394)
+(0.99879867, 'LITTLE')
 ```
 
 * Detection

ppocr/data/simple_dataset.py

@@ -74,7 +74,7 @@ class SimpleDataSet(Dataset):
         data_line = self.data_lines[file_idx]
         try:
             data_line = data_line.decode('utf-8')
-            substr = data_line.strip("\n").split(self.delimiter)
+            substr = data_line.strip("\n").strip("\r").split(self.delimiter)
             file_name = substr[0]
             label = substr[1]
             img_path = os.path.join(self.data_dir, file_name)

ppocr/modeling/heads/self_attention.py

@@ -285,8 +285,7 @@ class PrePostProcessLayer(nn.Layer):
             elif cmd == "n":  # add layer normalization
                 self.functors.append(
                     self.add_sublayer(
-                        "layer_norm_%d" % len(
-                            self.sublayers(include_sublayers=False)),
+                        "layer_norm_%d" % len(self.sublayers()),
                         paddle.nn.LayerNorm(
                             normalized_shape=d_model,
                             weight_attr=fluid.ParamAttr(
@@ -320,9 +319,7 @@ class PrepareEncoder(nn.Layer):
         self.src_emb_dim = src_emb_dim
         self.src_max_len = src_max_len
         self.emb = paddle.nn.Embedding(
-            num_embeddings=self.src_max_len,
-            embedding_dim=self.src_emb_dim,
-            sparse=True)
+            num_embeddings=self.src_max_len, embedding_dim=self.src_emb_dim)
         self.dropout_rate = dropout_rate
 
     def forward(self, src_word, src_pos):

ppocr/utils/save_load.py

@@ -80,9 +80,9 @@ def init_model(config, model, logger, optimizer=None, lr_scheduler=None):
     """
     load model from checkpoint or pretrained_model
     """
-    gloabl_config = config['Global']
-    checkpoints = gloabl_config.get('checkpoints')
-    pretrained_model = gloabl_config.get('pretrained_model')
+    global_config = config['Global']
+    checkpoints = global_config.get('checkpoints')
+    pretrained_model = global_config.get('pretrained_model')
     best_model_dict = {}
     if checkpoints:
         assert os.path.exists(checkpoints + ".pdparams"), \
@@ -105,7 +105,7 @@ def init_model(config, model, logger, optimizer=None, lr_scheduler=None):
 
         logger.info("resume from {}".format(checkpoints))
     elif pretrained_model:
-        load_static_weights = gloabl_config.get('load_static_weights', False)
+        load_static_weights = global_config.get('load_static_weights', False)
         if not isinstance(pretrained_model, list):
             pretrained_model = [pretrained_model]
         if not isinstance(load_static_weights, list):