Update app.py

app.py

@@ -76,9 +76,10 @@ class RAGChatbot:
         full_prompt = f"Context: {truncated_context}\n\nQuestion: {query}\n\nAnswer:"
         
         # Generate response
-        inputs = self.tokenizer(full_prompt, return_tensors="pt", padding=True, truncation=True).to(self.model.device)
-        attention_mask = inputs.attention_mask
-        outputs = self.model.generate(**inputs, max_new_tokens=128,attention_mask=attention_mask,pad_token_id=self.tokenizer.eos_token_id)
+        tokens = self.tokenizer(full_prompt, return_tensors="pt", padding=True, truncation=True).to(self.model.device)
+        inputs = tokens.input_ids.to(self.model.device)
+        attention_mask = tokens.attention_mask
+        outputs = self.model.generate(inputs, max_new_tokens=128,attention_mask=attention_mask,pad_token_id=self.tokenizer.eos_token_id,repetition_penalty=1.0))
         response = self.tokenizer.decode(outputs[0], skip_special_tokens=True)
         
         return response.split("Answer:")[-1].strip()