add need_reset to dataset

ppocr/data/lmdb_dataset.py

@@ -38,6 +38,9 @@ class LMDBDataSet(Dataset):
             np.random.shuffle(self.data_idx_order_list)
         self.ops = create_operators(dataset_config['transforms'], global_config)
 
+        ratio_list = dataset_config.get("ratio_list", [1.0])
+        self.need_reset = True in [x < 1 for x in ratio_list]
+
     def load_hierarchical_lmdb_dataset(self, data_dir):
         lmdb_sets = {}
         dataset_idx = 0

ppocr/data/pgnet_dataset.py

@@ -49,6 +49,8 @@ class PGDataSet(Dataset):
 
         self.ops = create_operators(dataset_config['transforms'], global_config)
 
+        self.need_reset = True in [x < 1 for x in ratio_list]
+
     def shuffle_data_random(self):
         if self.do_shuffle:
             random.seed(self.seed)

ppocr/data/pubtab_dataset.py

@@ -53,6 +53,9 @@ class PubTabDataSet(Dataset):
             self.shuffle_data_random()
         self.ops = create_operators(dataset_config['transforms'], global_config)
 
+        ratio_list = dataset_config.get("ratio_list", [1.0])
+        self.need_reset = True in [x < 1 for x in ratio_list]
+
     def shuffle_data_random(self):
         if self.do_shuffle:
             random.seed(self.seed)
@@ -85,7 +88,11 @@ class PubTabDataSet(Dataset):
                 cells = info['html']['cells'].copy()
                 structure = info['html']['structure'].copy()
                 img_path = os.path.join(self.data_dir, file_name)
-                data = {'img_path': img_path, 'cells': cells, 'structure':structure}
+                data = {
+                    'img_path': img_path,
+                    'cells': cells,
+                    'structure': structure
+                }
                 if not os.path.exists(img_path):
                     raise Exception("{} does not exist!".format(img_path))
                 with open(data['img_path'], 'rb') as f: