updates

app.py

@@ -42,16 +42,16 @@ class VideoAnnotator:
             logger.info(f"Loading dataset from HuggingFace: {HF_REPO_ID}")
             self.dataset = load_dataset(HF_REPO_ID, token=self.hf_token)
 
-            # Get the split (usually 'train')
+            # Usually "train" split
             split = list(self.dataset.keys())[0]
             self.dataset_split = self.dataset[split]
 
-            # Get all potential video files from the dataset
+            # Get all the video files from the CricBench dataset
             all_video_files = [item['video'] if 'video' in item else item['path']
                                 for item in self.dataset_split]
             logger.info(f"Found {len(all_video_files)} potential video files in source dataset.")
 
-            # --- Start Edit: Filter out already annotated videos ---
+            # Filter out already annotated videos
             logger.info(f"Checking for existing annotations in {self.annotation_repo_id}")
             try:
                 # List files in the 'annotations' directory of the annotation repo
@@ -65,7 +65,7 @@ class VideoAnnotator:
                 annotated_video_basenames = set(
                     os.path.basename(f).replace('.jsonl', '')
                     for f in annotated_files
-                    if f.startswith("annotations/") and f.endswith(".jsonl") # Ensure it's in the correct folder and has the right extension
+                    if f.startswith("annotations/") and f.endswith(".jsonl") 
                 )
                 logger.info(f"Found {len(annotated_video_basenames)} existing annotation files.")