Update app.py

app.py

@@ -3,13 +3,9 @@ from PIL import Image
 import torch
 from transformers import AutoModelForCausalLM, AutoTokenizer
 
-#Setting device to cuda
-torch.set_default_device("cuda")
 
 # # Ensure GPU usage if available
-# device = "cuda" if torch.cuda.is_available() else "cpu"
-
-torch.set_default_dtype()('torch.cuda.FloatTensor')
+device = "cuda" if torch.cuda.is_available() else "cpu"
 
 
 
@@ -25,7 +21,7 @@ def predict_answer(image, question, max_tokens=100):
     text = f"A chat between a curious user and an artificial intelligence assistant. The assistant gives helpful, detailed, and polite answers to the user's questions. USER: <image>\n{question}? ASSISTANT:"
     image = image.convert("RGB")
     
-    input_ids = tokenizer(text, return_tensors='pt').input_ids.to('cuda')
+    input_ids = tokenizer(text, return_tensors='pt').input_ids.to(device)
     image_tensor = model.image_preprocess(image)
     
     #Generate the answer