Update app.py

app.py

@@ -12,61 +12,58 @@ DEFAULT_MAX_NEW_TOKENS = 1024
 MAX_INPUT_TOKEN_LENGTH = int(os.getenv("MAX_INPUT_TOKEN_LENGTH", "4096"))
 
 DESCRIPTION = """\
-
-ZhongJing-2-1_8b Chat
-This Space demonstrates the ZhongJing-2-1_8b model, a fine-tuned model for chat instructions. Feel free to play with it, or duplicate to run generations without a queue! If you want to run your own service, you can also deploy the model on Inference Endpoints.
+仲景GPT-V2-1.8B
+博极医源,精勤不倦。Unlocking the Wisdom of Traditional Chinese Medicine with AI.
 """
 
 LICENSE = """
-
 <p/>
 ---
-As a derivate work of [ZhongJing-2-1_8b](https://huggingface.co/CMLM/ZhongJing-2-1_8b) by 医哲未来 of Fudan University, this demo is governed by the original license and acceptable use policy.
+This demo is governed by the original licenses of [ZhongJing-2-1_8b](https://huggingface.co/CMLM/ZhongJing-2-1_8b) and [Qwen1.5-1.8B-Chat](https://huggingface.co/Qwen/Qwen1.5-1.8B-Chat).
 """
 
-if not torch.cuda.is_available():
-    DESCRIPTION += "\n<p>Running on CPU 🥶 This demo does not work on CPU.</p>"
-
-if torch.cuda.is_available():
-    base_model_id = "Qwen/Qwen1.5-1.8B-Chat"
-    peft_model_id = "CMLM/ZhongJing-2-1_8b"
-    model = AutoModelForCausalLM.from_pretrained(base_model_id, torch_dtype=torch.float16, device_map="auto")
-    model.load_adapter(peft_model_id)
-    tokenizer = AutoTokenizer.from_pretrained(
-        peft_model_id,
-        padding_side="right",
-        trust_remote_code=True,
-        pad_token=''
-    )
+peft_model_id = "CMLM/ZhongJing-2-1_8b"
+base_model_id = "Qwen/Qwen1.5-1.8B-Chat"
+model = AutoModelForCausalLM.from_pretrained(base_model_id, device_map="auto")
+model.load_adapter(peft_model_id)
+tokenizer = AutoTokenizer.from_pretrained(
+    "CMLM/ZhongJing-2-1_8b",
+    padding_side="right", 
+    trust_remote_code=True,
+    pad_token=''
+)
 
-@spaces.GPU
+@spaces.gpu()
 def generate(
     message: str,
-    chat_history: list[tuple[str, str]],
-    system_prompt: str,
     max_new_tokens: int = 1024,
     temperature: float = 0.6,
-    top_p: float = 0.9,
+    top_p: float = 0.9, 
     top_k: int = 50,
     repetition_penalty: float = 1.2,
 ) -> Iterator[str]:
-    conversation = []
-    if system_prompt:
-        conversation.append({"role": "system", "content": system_prompt})
-    for user, assistant in chat_history:
-        conversation.extend([{"role": "user", "content": user}, {"role": "assistant", "content": assistant}])
-    conversation.append({"role": "user", "content": message})
 
-    input_ids = tokenizer.apply_chat_template(conversation, return_tensors="pt")
+    prompt = f"Question: {message}"
+    messages = [
+        {"role": "system", "content": "You are a helpful TCM medical assistant named 仲景中医大语言模型, created by 医哲未来 of Fudan University."},
+        {"role": "user", "content": prompt}
+    ]
+
+    text = tokenizer.apply_chat_template(
+        messages,
+        tokenize=False,
+        add_generation_prompt=True
+    )
+    input_ids = tokenizer([text], return_tensors="pt").input_ids
+
     if input_ids.shape[1] > MAX_INPUT_TOKEN_LENGTH:
         input_ids = input_ids[:, -MAX_INPUT_TOKEN_LENGTH:]
         gr.Warning(f"Trimmed input from conversation as it was longer than {MAX_INPUT_TOKEN_LENGTH} tokens.")
-    input_ids = input_ids.to(model.device)
 
     streamer = TextIteratorStreamer(tokenizer, timeout=10.0, skip_prompt=True, skip_special_tokens=True)
     generate_kwargs = dict(
         input_ids=input_ids,
-        streamer=streamer,
+        streamer=streamer, 
         max_new_tokens=max_new_tokens,
         do_sample=True,
         top_p=top_p,
@@ -83,54 +80,55 @@ def generate(
         outputs.append(text)
         yield "".join(outputs)
 
-chat_interface = gr.ChatInterface(
+chat_interface = gr.Interface(
     fn=generate,
-    additional_inputs=[
-        gr.Textbox(label="System prompt", lines=6),
-        gr.Slider(
+    inputs=[
+        gr.components.Textbox(label="Enter your question"),
+        gr.components.Slider(
             label="Max new tokens",
-            minimum=1,
+            minimum=1,  
             maximum=MAX_MAX_NEW_TOKENS,
             step=1,
             value=DEFAULT_MAX_NEW_TOKENS,
         ),
-        gr.Slider(
+        gr.components.Slider( 
             label="Temperature",
             minimum=0.1,
             maximum=4.0,
-            step=0.1,
+            step=0.1, 
             value=0.6,
         ),
-        gr.Slider(
+        gr.components.Slider(
             label="Top-p (nucleus sampling)",
             minimum=0.05,
             maximum=1.0,
             step=0.05,
             value=0.9,
         ),
-        gr.Slider(
+        gr.components.Slider(
             label="Top-k",
             minimum=1,
             maximum=1000,
             step=1,
             value=50,
         ),
-        gr.Slider(
-            label="Repetition penalty",
+        gr.components.Slider(
+            label="Repetition penalty", 
             minimum=1.0,
             maximum=2.0,
             step=0.05,
             value=1.2,
         ),
     ],
-    stop_btn=None,
+    outputs="text",
+    title="仲景GPT-V2-1.8B",
+    description=DESCRIPTION,
+    allow_flagging=False,
     examples=[
-        ["Hello there! How are you doing?"],
-        ["Can you explain briefly to me what is the Python programming language?"],
-        ["Explain the plot of Cinderella in a sentence."],
-        ["How many hours does it take a man to eat a Helicopter?"],
-        ["Write a 100-word article on 'Benefits of Open-Source in AI research'"],
-    ],
+        ["请问气虚体质有哪些症状表现?"],
+        ["简单介绍一下中医的五行学说。"],
+        ["桑螵蛸是什么?有什么功效作用?"],
+    ],  
 )
 
 with gr.Blocks(css="style.css") as demo:
@@ -140,4 +138,4 @@ with gr.Blocks(css="style.css") as demo:
     gr.Markdown(LICENSE)
 
 if __name__ == "__main__":
-    demo.queue(max_size=20).launch()
+    demo.queue(max_size=20).launch()