!3729 add create_train_lst.py for PSPNet

Merge pull request !3729 from hemaohua/master

research/cv/PSPNet/README.md

@@ -37,6 +37,12 @@ The pyramid pooling module fuses features under four different pyramid scales.Fo
 
 - [PASCAL VOC 2012 and SBD Dataset Website](http://home.bharathh.info/pubs/codes/SBD/download.html)
  - It contains 11,357 finely annotated images split into training and testing sets with 8,498 and 2,857 images respectively.
+ - The path formats in voc_train_lst.txt and voc_val_lst.txt are different, you can run create_train_lst.py to generate train_lst.txt in data dir for VOC2012. As follow：
+
+ ```python
+ python src/dataset/create_train_lst.py --data_dir [DATA_DIR]
+ ```
+
 - [ADE20K Dataset Website](http://groups.csail.mit.edu/vision/datasets/ADE20K/)
  - It contains 22,210 finely annotated images split into training and testing sets with 20,210 and 2,000 images respectively.
  - The ADE20k directory structure is as follows:

research/cv/PSPNet/src/dataset/create_train_lst.py

+# Copyright 2022 Huawei Technologies Co., Ltd
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+# ============================================================================
+"""generate train_lst.txt"""
+import os
+import argparse
+
+
+def _parser_args():
+    parser = argparse.ArgumentParser('dataset list generator')
+    parser.add_argument("--data_dir", type=str, default='', help="VOC2012 data dir")
+    return parser.parse_args()
+
+
+def _get_data_list(data_list_file):
+    with open(data_list_file, 'r') as f:
+        return f.readlines()
+
+
+def main():
+    args = _parser_args()
+    data_dir = args.data_dir
+    voc_train_lst_txt = os.path.join(data_dir, 'voc_train_lst.txt')
+    train_lst_txt = os.path.join(data_dir, 'train_lst.txt')
+
+    voc_train_data_lst = _get_data_list(voc_train_lst_txt)
+    with open(train_lst_txt, 'w') as f:
+        for line in voc_train_data_lst:
+            img_, anno_ = (os.path.join('VOCdevkit/VOC2012', i.strip()) for i in line.split())
+            f.write(f'{img_} {anno_}\n')
+    print('generating voc train list success.')
+
+
+if __name__ == "__main__":
+    main()