Update app.py

app.py

@@ -16,7 +16,7 @@ def query(instruction, input):
         {input}
         ### Response:
         """
-    input_ids = tokenizer(prompt, return_tensors="pt", truncation=True).input_ids.cuda()
+    input_ids = tokenizer(prompt, return_tensors="pt", truncation=True)
     output_base = base_model.generate(input_ids=input_ids, max_new_tokens=500, do_sample=True, top_p=0.9,temperature=0.5)
     response = "{tokenizer.batch_decode(output_base.detach().cpu().numpy(), skip_special_tokens=True)[0][len(prompt):]}"
     return response