Update app.py

app.py

@@ -23,24 +23,6 @@ def run_commands():
 # Call the function to run the commands
 
 if __name__ == "__main__":
-    run_commands()
-
-    from typing import List
-    from Utils import get_video_properties
-    from GroundingDINO.groundingdino.util.inference import load_model, predict
-    import cv2
-    import numpy as np
-    import torch
-    from PIL import Image
-    import GroundingDINO.groundingdino.datasets.transforms as T
-    from torchvision.ops import box_convert
-    from torchvision import transforms
-    from torch import nn
-    from os.path import dirname, abspath
-    import yaml
-    import supervision as sv
-    import gradio as gr
-    import spaces
 
     class DinoVisionTransformerClassifier(nn.Module):
         def __init__(self):
@@ -167,6 +149,24 @@ if __name__ == "__main__":
 
         @spaces.GPU(duration=30)
         def generate_video(self, video_path) -> None:
+            run_commands()
+
+            from typing import List
+            from Utils import get_video_properties
+            from GroundingDINO.groundingdino.util.inference import load_model, predict
+            import cv2
+            import numpy as np
+            import torch
+            from PIL import Image
+            import GroundingDINO.groundingdino.datasets.transforms as T
+            from torchvision.ops import box_convert
+            from torchvision import transforms
+            from torch import nn
+            from os.path import dirname, abspath
+            import yaml
+            import supervision as sv
+            import gradio as gr
+            import spaces
 
             # Load model, set up variables and get video properties
             cap, fps, width, height, fourcc = get_video_properties(video_path)