Update CC6204-Hackaton-Cub-Dataset.py

CC6204-Hackaton-Cub-Dataset.py

@@ -112,18 +112,17 @@ class CubDataset(datasets.GeneratorBasedBuilder):
        
       for img, text in zip(img_path_files, text_path_files):
       #for img in img_path_files:
-         print(img)
-         print(os.path.basename(img))
-         print(text)
-         print(os.path.basename(text))
+         #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)]
-         if img_idx in _TRAIN_IDX_SET:
-            train_files.append((img, text))
-            train_idx.append(img_idx)
-         else:
-            test_files.append((img, text))
-            test_idx.append(img_idx)
+         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))
+               train_idx.append(img_idx)
+            else:
+               test_files.append((img, text))
+               test_idx.append(img_idx)
                
       return [
                  datasets.SplitGenerator(