Update app_chat.py

app_chat.py

@@ -10,8 +10,6 @@ from transformers import StoppingCriteria, StoppingCriteriaList, StopStringCrite
 
 import subprocess
 
-global model
-
 import torch._dynamo
 torch._dynamo.config.suppress_errors = True
 
@@ -58,8 +56,6 @@ def generate(
     repetition_penalty: float = 1.2,
 ) -> Iterator[str]:
     conversation = []
-
-    global model
     
     if system_prompt:
         conversation.append({"role": "system", "content": system_prompt})
@@ -96,14 +92,10 @@ def generate(
     t.start()
 
     outputs = []
-    try:
-        for text in streamer:
-            outputs.append(text)
-            yield "".join(outputs)
-    except:
-        print("restarting the model, got some error")
-        model = AutoModelForCausalLM.from_pretrained(model_id, device_map="cuda", trust_remote_code=True)
-        model = model.cuda().to(torch.bfloat16)
+    for text in streamer:
+        outputs.append(text)
+        yield "".join(outputs)
+
 
 chat_interface = gr.ChatInterface(
     fn=generate,