Update CC6204-Hackaton-Cub-Dataset.py

CC6204-Hackaton-Cub-Dataset.py

@@ -5,8 +5,8 @@ import datasets
 
 from requests import get
 
-datasets.logging.set_verbosity_debug()
-logger = datasets.logging.get_logger(__name__)
+#datasets.logging.set_verbosity_debug()
+#logger = datasets.logging.get_logger(__name__)
 #datasets.logging.set_verbosity_info()
 #datasets.logging.set_verbosity_debug()
 
@@ -112,11 +112,8 @@ class CubDataset(datasets.GeneratorBasedBuilder):
       text_path_files = dl_manager.iter_files(text_data_files)
        
       for img, text in zip(img_path_files, text_path_files):
-      #for img in img_path_files:
-         #print(os.path.basename(img))
-         #print(os.path.basename(text))
-         #text = text_path_files[text_path_files.index(os.path.basename(img).replace("jpg", "txt"))]
          img_idx = _IMGNAME2ID[os.path.basename(img)]
+         # Sanity check to ensure that pairs of text and image are correct
          if os.path.basename(img).replace(".jpg", "") == os.path.basename(text).replace(".txt", ""):
             if img_idx in _TRAIN_IDX_SET:
                train_files.append((img, text))