Update app.py

app.py

@@ -1,64 +1,51 @@
 import gradio as gr
-from huggingface_hub import InferenceClient
-
-"""
-For more information on `huggingface_hub` Inference API support, please check the docs: https://huggingface.co/docs/huggingface_hub/v0.22.2/en/guides/inference
-"""
-client = InferenceClient("HuggingFaceH4/zephyr-7b-beta")
-
-
-def respond(
-    message,
-    history: list[tuple[str, str]],
-    system_message,
-    max_tokens,
-    temperature,
-    top_p,
-):
-    messages = [{"role": "system", "content": system_message}]
-
-    for val in history:
-        if val[0]:
-            messages.append({"role": "user", "content": val[0]})
-        if val[1]:
-            messages.append({"role": "assistant", "content": val[1]})
-
-    messages.append({"role": "user", "content": message})
-
-    response = ""
+from transformers import AutoTokenizer, AutoModelForCausalLM
+import torch
+import spaces
+
+
+model_name = "Zhihu-ai/Zhi-writing-dsr1-14"
+tokenizer = AutoTokenizer.from_pretrained(model_name, trust_remote_code=True)
+model = AutoModelForCausalLM.from_pretrained(
+    model_name,
+    torch_dtype=torch.float16,
+    device_map="auto",
+    trust_remote_code=True
+)
 
-    for message in client.chat_completion(
-        messages,
-        max_tokens=max_tokens,
-        stream=True,
-        temperature=temperature,
-        top_p=top_p,
+@spaces.GPU()
+def predict(message, history):
+    
+    history_text = ""
+    for human, assistant in history:
+        history_text += f"Human: {human}\nAssistant: {assistant}\n"
+    prompt = f"{history_text}Human: {message}\nAssistant:"
+    
+    # 生成回复
+    inputs = tokenizer(prompt, return_tensors="pt").to(model.device)
+    
+    # 使用流式生成
+    for response in model.generate(
+        **inputs,
+        max_new_tokens=10000,
+        do_sample=True,
+        temperature=0.7,
+        top_p=0.9,
+        repetition_penalty=1.1,
+        pad_token_id=tokenizer.eos_token_id,
+        streamer=gr.TextStreamer(tokenizer, skip_prompt=True, skip_special_tokens=True)
     ):
-        token = message.choices[0].delta.content
-
-        response += token
-        yield response
+        yield response.strip()
 
-
-"""
-For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface
-"""
+# 创建Gradio界面
 demo = gr.ChatInterface(
-    respond,
-    additional_inputs=[
-        gr.Textbox(value="You are a friendly Chatbot.", label="System message"),
-        gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens"),
-        gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature"),
-        gr.Slider(
-            minimum=0.1,
-            maximum=1.0,
-            value=0.95,
-            step=0.05,
-            label="Top-p (nucleus sampling)",
-        ),
-    ],
+    predict,
+    title="Zhi-writing-dsr1-14",
+    description="这是一个基于Zhi-writing-dsr1-14的文章生成器。",
+    examples=["以鲁迅口吻写一篇500字关于桔了个仔的散文", "用知乎常见的表达方式讲讲什么是AI？", "告诉我一个我大概率不知道的人生哲理"],
+    theme=gr.themes.Soft(),
+    streaming=True
 )
 
-
 if __name__ == "__main__":
-    demo.launch()
+    demo.launch(share=True)