Update app.py

app.py

@@ -72,7 +72,7 @@ class DocumentRetrievalAndGeneration:
         return generate_text
     def initialize_llm2(self,model_id):
         
-        client = InferenceClient("HuggingFaceH4/zephyr-7b-beta")
+        self.client = InferenceClient("HuggingFaceH4/zephyr-7b-beta")
         # except:
         #     try:
         #         pipe = pipeline("text-generation", model="microsoft/Phi-3-mini-4k-instruct", trust_remote_code=True)
@@ -156,7 +156,7 @@ class DocumentRetrievalAndGeneration:
         messages.append({"role": "user", "content": query})
         response = ""
 
-        for message in client.chat_completion(messages,max_tokens=2048,stream=True,temperature=0.7):
+        for message in self.client.chat_completion(messages,max_tokens=2048,stream=True,temperature=0.7):
             token = message.choices[0].delta.content
             response += token
         # yield response