Add text description

CC6204-Hackaton-Cub-Dataset.py

@@ -98,19 +98,24 @@ class CubDataset(datasets.GeneratorBasedBuilder):
    def _split_generators(self, dl_manager):      
       train_files = []
       train_idx = []
+       
       test_files = []
       test_idx = []
       
       # Download images
-      data_files = dl_manager.download_and_extract(_URLS["image_urls"])
-      path_files = dl_manager.iter_files(data_files)
-      for img in path_files:
+      img_data_files = dl_manager.download_and_extract(_URLS["image_urls"])
+      text_data_files = dl_manager.download_and_extract(_URLS["text_urls"])
+       
+      img_path_files = dl_manager.iter_files(img_data_files)
+      text_path_files = dl_manager.iter_files(text_data_files)
+       
+      for img, text in zip(img_path_files, text_path_files):
          img_idx = _IMGNAME2ID[os.path.basename(img)]
          if img_idx in _TRAIN_IDX_SET:
-            train_files.append(img)
+            train_files.append((img, text))
             train_idx.append(img_idx)
          else:
-            test_files.append(img)
+            test_files.append((img, text))
             test_idx.append(img_idx)
                
       return [
@@ -134,11 +139,11 @@ class CubDataset(datasets.GeneratorBasedBuilder):
    def _generate_examples(self, files, image_idx):
    
       for i, path in enumerate(files):
-         file_name = os.path.basename(path)
+         file_name = os.path.basename(path[0])
          if file_name.endswith(".jpg"):
             yield i, {
-               "image": path,
-               "description": "lala",
+               "image": path[0],
+               "description": path[1],
                "label": _ID2LABEL[_IMGID2CLASSID[image_idx[i]]],
             }