Update app.py

app.py

@@ -37,13 +37,14 @@ h3 {
 
 device = "cuda" # for GPU usage or "cpu" for CPU usage
 
-tokenizer = AutoTokenizer.from_pretrained(MODEL)
+tokenizer = AutoTokenizer.from_pretrained(MODEL, use_fast=False, force_download=True)
 model = AutoModelForCausalLM.from_pretrained(
     MODEL,
     torch_dtype=torch.bfloat16,
     device_map="auto",
     trust_remote_code=True,
-    ignore_mismatched_sizes=True)
+    ignore_mismatched_sizes=True,
+    force_download=True)
 
 def format_chat(system_prompt, history, message):
     formatted_chat = f"<|im_start|>system\n{system_prompt}<|im_end|>\n"
@@ -60,10 +61,8 @@ def stream_chat(
     system_prompt: str,
     temperature: float = 0.3, 
     max_new_tokens: int = 256, 
-    top_p: float = 1.0
-, 
+    top_p: float = 1.0, 
     top_k: int = 20, 
-
     repetition_penalty: float = 1.2,
 ):
     print(f'message: {message}')
@@ -72,8 +71,7 @@ def stream_chat(
     formatted_prompt = format_chat(system_prompt, history, message)
     inputs = tokenizer(formatted_prompt, return_tensors="pt").to(device)
     
-    streamer = TextIteratorStreamer(tokenizer, timeout=5000.0
-, skip_prompt=True, skip_special_tokens=True)
+    streamer = TextIteratorStreamer(tokenizer, timeout=5000.0, skip_prompt=True, skip_special_tokens=True)
     
     generate_kwargs = dict(
         input_ids=inputs.input_ids,
@@ -167,4 +165,4 @@ with gr.Blocks(css=CSS, theme="soft") as demo:
     )
 
 if __name__ == "__main__":
-    demo.launch()
+    demo.launch()