placed timeout on streamer

app.py

@@ -53,7 +53,7 @@ def llama_generation(input_text: str,
     input_ids = llama_tokenizer.apply_chat_template(conversation, return_tensors='pt').to(llama_model.device)
 
     # Skip_prompt, ignores the prompt in the chatbot
-    streamer = TextIteratorStreamer(llama_tokenizer, skip_prompt=True, skip_special_tokens=True)
+    streamer = TextIteratorStreamer(llama_tokenizer, timeout=10.0, skip_prompt=True, skip_special_tokens=True)
 
     # generation arguments to pass in llm generate() eventually
     generate_kwargs = dict(
@@ -78,7 +78,7 @@ def llama_generation(input_text: str,
     for text in streamer:
         outputs.append(text)
         print(outputs)
-        return "".join(outputs)
+        yield "".join(outputs)
 
 # Let's just make sure the llama is returning as it should and than place that return output into a function making it fit into a base
 # Prompt for gpt-4o