Update app.py

app.py

@@ -2,8 +2,8 @@
 
 # gr.load("models/ManishThota/InstructBlip-VQA").launch()
 
-
 from PIL import Image
+import torch
 from transformers import BlipProcessor, BlipForQuestionAnswering
 
 # Initialize the model and processor
@@ -15,7 +15,7 @@ def predict_answer(image, question):
     image = image.convert("RGB")
 
     # Prepare inputs
-    encoding = processor(image, question, return_tensors="pt").to("cuda:0", torch.float16)
+    encoding = processor(image, question, return_tensors="pt")
 
     out = model.generate(**encoding)
     generated_text = processor.decode(out[0], skip_special_tokens=True)