Update app.py

app.py

@@ -11,13 +11,13 @@ parser = argparse.ArgumentParser()
 parser.add_argument("--cpu", action="store_true", help="Use CPU for computation")
 args = parser.parse_args([])
 
-DEVICE = "cpu"
-DTYPE = torch.float32 
+DEVICE = "cuda" if torch.cuda.is_available() and not args.cpu else "cpu" # Determine device based on availability and argument
+DTYPE = torch.float32 if DEVICE == "cpu" else torch.float16 # CPU doesn't support float16
 LATEST_REVISION = "2024-05-20"
-
+MODEL_ID = "yeshavyas27/moondream-ft"
 tokenizer = AutoTokenizer.from_pretrained("vikhyatk/moondream2", revision=LATEST_REVISION)
 moondream = AutoModelForCausalLM.from_pretrained(
-    "yeshavyas27/moondream-ft", trust_remote_code=True, torch_dtype=DTYPE
+    MODEL_ID, trust_remote_code=True, torch_dtype=DTYPE
 ).to(device=DEVICE)
 
 moondream.eval()
@@ -94,5 +94,4 @@ with gr.Blocks() as demo:
     prompt.submit(answer_question, [img, prompt], output)
     output.change(process_answer, [img, output], ann, show_progress=False)
 
-
-demo.queue().launch(debug=False)
+demo.queue().launch(debug=True)