Update app.py

app.py

@@ -11,18 +11,12 @@ import tempfile
 import shutil
 
 # Check if CUDA is available, otherwise use CPU
-device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
+device = "cuda"
 
 processor = Owlv2Processor.from_pretrained("google/owlv2-base-patch16")
 model = Owlv2ForObjectDetection.from_pretrained("google/owlv2-base-patch16")
 
-# Try to move model to GPU and use half precision
-try:
-    model = model.to(device).half()
-except RuntimeError:
-    print("GPU out of memory, using CPU instead")
-    device = torch.device("cpu")
-    model = model.to(device)
+model = model.to(device).half()
 
 @spaces.GPU(duration=120)
 def process_video(video_path, target, progress=gr.Progress()):