Update app.py

app.py

@@ -1,6 +1,5 @@
 from threading import Thread
 import gradio as gr
-import transformers
 import torch
 from transformers import (
     pipeline,
@@ -19,16 +18,16 @@ def chat_history(history) -> str:
         for i, dialog in enumerate(history) for _ in (0, 1) if dialog[i % 2]
     ]
 
-    return pipeline.tokenizer.apply_chat_template(
+    return pipe.tokenizer.apply_chat_template(
         messages, toknizer=False, add_generation_prompt=True
     )
 
 
 def model_loading_pipeline():
-    model_id = "vilm/vinallama-2.7b"
+    model_id = "vilm/vinallama-7b"
     tokenizer = AutoTokenizer.from_pretrained(model_id)
     streamer = TextIteratorStreamer(tokenizer, skip_prompt=True, Timeout=5)
-    pipeline = transformers.pipeline(
+    pipe = pipeline(
         "text-generation",
         model=model_id,
         model_kwargs={
@@ -36,10 +35,10 @@ def model_loading_pipeline():
         },
         streamer=streamer,
     )
-    return pipeline, streamer
+    return pipe, streamer
 
 
-def launch_app(pipeline, streamer):
+def launch_app(pipe, streamer):
     with gr.Blocks() as demo:
         chat = gr.Chatbot()
         msg = gr.Textbox()
@@ -59,7 +58,7 @@ def launch_app(pipeline, streamer):
                 "top_k": 50,
                 "top_p": 0.95,
             }
-            thread = Thread(target=pipeline, kwargs=kwargs)
+            thread = Thread(target=pipe, kwargs=kwargs)
             thread.start()
 
             for token in streamer: