add general_chars.txt

model_zoo/official/cv/crnn_seq2seq_ocr/scripts/run_eval_ascend.sh

@@ -55,6 +55,7 @@ then
 fi
 mkdir ./eval
 cp ../*.py ./eval
+cp ../*.txt ./eval
 cp *.sh ./eval
 cp -r ../src ./eval
 cd ./eval || exit

model_zoo/official/cv/crnn_seq2seq_ocr/src/config.py

@@ -28,6 +28,7 @@ config = ed({
     "val_annotation_file": "",
     "data_json": "",
 
+    "go_shift": 1,
     "characters_dictionary": {"pad_id": 0, "go_id": 1, "eos_id": 2, "unk_id": 3},
     "labels_not_use": [u'%#�?%', u'%#背景#%', u'%#不识�?%', u'#%不识�?#', u'%#模糊#%', u'%#模糊#%'],
     "vocab_path": "./general_chars.txt",

model_zoo/official/cv/crnn_seq2seq_ocr/src/create_mindrecord_files.py

@@ -19,8 +19,8 @@ import numpy as np
 
 from mindspore.mindrecord import FileWriter
 
-from src.config import config
-from src.utils import initialize_vocabulary
+from config import config
+from utils import initialize_vocabulary
 
 
 def serialize_annotation(img_path, lex, vocab):
@@ -82,7 +82,7 @@ def create_fsns_label(image_dir, anno_file_dirs):
 
 def fsns_train_data_to_mindrecord(mindrecord_dir, prefix="data_ocr.mindrecord", file_num=8):
 
-    anno_file_dirs = [config.train_annotation_file]
+    anno_file_dirs = [config.annotation_file]
     images, image_path_dict, image_anno_dict = create_fsns_label(image_dir=config.data_root,
                                                                  anno_file_dirs=anno_file_dirs)
     vocab, _ = initialize_vocabulary(config.vocab_path)
@@ -104,8 +104,8 @@ def fsns_train_data_to_mindrecord(mindrecord_dir, prefix="data_ocr.mindrecord",
         image_path = image_path_dict[img_id]
         annotation = image_anno_dict[img_id]
 
-        label_max_len = config.max_text_len
-        text_max_len = config.max_text_len - 2
+        label_max_len = config.max_length
+        text_max_len = config.max_length - 2
 
         if len(annotation) > text_max_len:
             continue
@@ -151,8 +151,8 @@ def fsns_train_data_to_mindrecord(mindrecord_dir, prefix="data_ocr.mindrecord",
 
 def fsns_val_data_to_mindrecord(mindrecord_dir, prefix="data_ocr.mindrecord", file_num=8):
 
-    anno_file_dirs = [config.train_annotation_file]
-    images, image_path_dict, image_anno_dict = create_fsns_label(image_dir=config.data_root,
+    anno_file_dirs = [config.val_annotation_file]
+    images, image_path_dict, image_anno_dict = create_fsns_label(image_dir=config.val_data_root,
                                                                  anno_file_dirs=anno_file_dirs)
     vocab, _ = initialize_vocabulary(config.vocab_path)
 
@@ -171,8 +171,8 @@ def fsns_val_data_to_mindrecord(mindrecord_dir, prefix="data_ocr.mindrecord", fi
         image_path = image_path_dict[img_id]
         annotation = image_anno_dict[img_id]
 
-        label_max_len = config.max_text_len
-        text_max_len = config.max_text_len - 2
+        label_max_len = config.max_length
+        text_max_len = config.max_length - 2
 
         if len(annotation) > text_max_len:
             continue
@@ -226,7 +226,7 @@ def create_mindrecord(dataset="fsns", prefix="fsns.mindrecord", is_training=True
                 print("{} dataset is not defined!".format(dataset))
 
     if not is_training:
-        mindrecord_dir = os.path.join(config.mindrecord_dir, "val")
+        mindrecord_dir = os.path.join(config.mindrecord_dir, "test")
         mindrecord_files = [os.path.join(mindrecord_dir, prefix + "0")]
 
         if not os.path.exists(mindrecord_files[0]):
@@ -243,3 +243,9 @@ def create_mindrecord(dataset="fsns", prefix="fsns.mindrecord", is_training=True
                 print("{} dataset is not defined!".format(dataset))
 
     return mindrecord_files
+
+
+if __name__ == '__main__':
+    create_mindrecord(is_training=True)
+    create_mindrecord(is_training=False)
+    print("END")