Update utils.py

utils.py

@@ -352,7 +352,7 @@ def rag_chain(llm, prompt,  retriever):
         inputs = tokenizer_rag(input_text, return_tensors="pt", max_length=1024, truncation=True)
 
         #Generiere die Antwort
-        outputs = model_rag.generate(inputs['input_ids'], max_length=150, num_beams=2, early_stopping=True)
+        outputs = modell_rag.generate(inputs['input_ids'], max_length=150, num_beams=2, early_stopping=True)
         answer = tokenizer_rag.decode(outputs[0], skip_special_tokens=True)