Update app.py

app.py

@@ -11,28 +11,29 @@ model = AutoModelForCausalLM.from_pretrained(model_name, trust_remote_code=True)
 tokenizer = AutoTokenizer.from_pretrained(model_name, trust_remote_code=True)
 
 # 生成回复的函数
-def chat(input_text, history=[]):
-    inputs = tokenizer(input_text, return_tensors="pt").to(device)
+def chat(input_text, chat_history=[]):
+    inputs = tokenizer(input_text, return_tensors="pt").to("cuda")
     with torch.no_grad():
-        outputs = model.generate(**inputs, max_length=512, temperature=0.7)
+        outputs = model.generate(**inputs, max_length=512)
     response = tokenizer.decode(outputs[0], skip_special_tokens=True)
-    return response
+    chat_history.append((input_text, response))
+    return chat_history, ""
 
 # 创建 Gradio 界面
 with gr.Blocks() as demo:
-    gr.Markdown("## DeepSeek-V3 Chatbot")
-    
+    gr.Markdown("# 🤖 Chatbot")
     chatbot = gr.Chatbot(label="Chatbot")
-    msg = gr.Textbox(label="Type a message...", placeholder="Enter your message here...", lines=2)
-    submit = gr.Button("Submit")
+    msg = gr.Textbox(placeholder="Type a message...")
+    with gr.Row():
+        submit_btn = gr.Button("Submit")
+        retry_btn = gr.Button("🔄 Retry")
+        undo_btn = gr.Button("↩️ Undo")
+        clear_btn = gr.Button("🗑️ Clear")
 
-    def respond(user_input, chat_history):
-        response = chat(user_input, chat_history)
-        chat_history.append((user_input, response))
-        return "", chat_history
+    submit_btn.click(chat, inputs=[msg, chatbot], outputs=[chatbot, msg])
+    retry_btn.click(chat, inputs=[msg, chatbot], outputs=[chatbot, msg])
+    undo_btn.click(lambda history: history[:-1] if history else [], inputs=[chatbot], outputs=[chatbot])
+    clear_btn.click(lambda: [], inputs=[], outputs=[chatbot])
 
-    submit.click(respond, inputs=[msg, chatbot], outputs=[msg, chatbot])
-
-# 启动 Gradio 应用
-if __name__ == "__main__":
-    demo.launch()
+# 运行 Gradio 界面
+demo.launch()