Update app.py

app.py

@@ -6,10 +6,10 @@ LoongFlow HuggingFace Spaces Demo
 """
 
 import gradio as gr
-import json
+import pandas as pd
 import time
 import random
-from typing import List, Dict, Any
+from typing import List, Dict, Any, Tuple
 
 # ============================================================================
 # PES 工作流程模拟
@@ -98,7 +98,7 @@ def solve_task(task):
         "role": "Executor",
         "action": "编写并执行代码",
         "code": code,
-        "result": "代码执行成功，所有测试通过 ✓",
+        "result": "代码执行成功，所有测试通过",
         "timestamp": time.strftime("%H:%M:%S")
     }
 
@@ -132,14 +132,14 @@ def simulate_summary(iteration: int, score: float) -> Dict[str, Any]:
     }
 
 
-def run_pes_iteration(task: str, iteration: int, current_score: float) -> List[Dict[str, Any]]:
+def run_pes_iteration(task: str, iteration: int, current_score: float) -> Tuple[List[Dict[str, Any]], float]:
     """运行一次完整的 PES 迭代"""
     results = []
     
     # Phase 1: Plan
     planner_result = simulate_planner(task)
     results.append({
-        "phase": "📋 Plan (计划)",
+        "phase": "Plan (计划)",
         "content": planner_result["thought"],
         "detail": planner_result["plan"],
         "timestamp": planner_result["timestamp"]
@@ -148,18 +148,18 @@ def run_pes_iteration(task: str, iteration: int, current_score: float) -> List[D
     # Phase 2: Execute
     executor_result = simulate_executor(task, planner_result["plan"])
     results.append({
-        "phase": "⚡ Execute (执行)",
+        "phase": "Execute (执行)",
         "content": executor_result["action"],
-        "detail": f"{executor_result['code']}\n\n**执行结果**: {executor_result['result']}",
+        "detail": f"{executor_result['code']}\n\n执行结果: {executor_result['result']}",
         "timestamp": executor_result["timestamp"]
     })
     
     # Phase 3: Summary
     summary_result = simulate_summary(iteration, current_score)
     results.append({
-        "phase": "📊 Summary (总结)",
+        "phase": "Summary (总结)",
         "content": summary_result["reflection"],
-        "detail": f"**改进建议**: {summary_result['improvement']}\n\n**当前分数**: {summary_result['score']:.2f}",
+        "detail": f"改进建议: {summary_result['improvement']}\n\n当前分数: {summary_result['score']:.2f}",
         "timestamp": summary_result["timestamp"]
     })
     
@@ -175,7 +175,7 @@ def create_demo():
     
     with gr.Blocks(title="LoongFlow PES Demo", theme=gr.themes.Soft()) as demo:
         gr.Markdown("""
-        # 🐉 LoongFlow PES Agent Demo
+        # LoongFlow PES Agent Demo
         
         **LoongFlow** 是一个进化式 Agent 开发框架，采用 **PES (Plan-Execute-Summary)** 思考范式。
         
@@ -186,16 +186,10 @@ def create_demo():
         ### PES 工作流程
         
         ```
-        ┌─────────────┐     ┌─────────────┐     ┌─────────────┐
-        │    Plan     │────▶│   Execute   │────▶│   Summary   │
-        │   (计划)    │     │   (执行)    │     │   (总结)    │
-        └─────────────┘     └─────────────┘     └─────────────┘
-              │                                           │
-              │              迭代改进 ◀───────────────────┘
-              │                    
-        ┌─────────────┐           
-        │   目标达成   │           
-        └─────────────┘           
+        Plan (计划) --> Execute (执行) --> Summary (总结)
+              ^                                    |
+              |____________________________________|
+                          迭代改进
         ```
         """)
         
@@ -217,85 +211,75 @@ def create_demo():
                         label="目标分数"
                     )
                 
-                run_btn = gr.Button("🚀 开始执行任务", variant="primary")
+                run_btn = gr.Button("开始执行任务", variant="primary")
             
             with gr.Column(scale=1):
-                status_output = gr.JSON(
+                status_output = gr.Textbox(
                     label="执行状态",
-                    visible=False
+                    lines=5,
+                    interactive=False
                 )
         
-        with gr.Row():
-            score_chart = gr.LinePlot(
-                label="分数演进",
-                x="iteration",
-                y="score",
-                height=300
-            )
+        score_chart = gr.LinePlot(
+            label="分数演进",
+            x="iteration",
+            y="score",
+            height=300
+        )
         
-        with gr.Accordion("📝 迭代详情", open=True):
-            iteration_output = gr.JSON(
+        with gr.Accordion("迭代详情", open=True):
+            iteration_output = gr.Textbox(
                 label="迭代历史",
-                elem_id="iteration-output"
+                lines=15,
+                interactive=False
             )
         
-        # 状态变量
-        iterations_data = []
-        
         def run_task(task: str, max_iter: int, target: float):
-            if not task.strip():
-                yield {"error": "请输入任务描述"}, [], gr.update()
+            if not task or not task.strip():
+                yield "错误: 请输入任务描述", pd.DataFrame(columns=["iteration", "score"]), ""
                 return
             
-            iterations_data.clear()
-            yield {"status": "开始执行任务...", "task": task}, [], gr.update()
-            
+            all_results_text = ""
+            chart_data = []
             current_score = 0.0
-            all_results = []
             
-            for i in range(1, max_iter + 1):
-                yield {
-                    "status": f"正在执行第 {i} 次迭代...",
-                    "iteration": i,
-                    "score": current_score
-                }, [], gr.update(visible=True)
+            yield "状态: 开始执行任务...", pd.DataFrame(columns=["iteration", "score"]), ""
+            
+            for i in range(1, int(max_iter) + 1):
+                status = f"状态: 正在执行第 {i} 次迭代..."
+                yield status, pd.DataFrame(chart_data) if chart_data else pd.DataFrame(columns=["iteration", "score"]), all_results_text
                 
                 results, current_score = run_pes_iteration(task, i, current_score)
-                all_results.extend(results)
+                
+                # 构建迭代结果文本
+                all_results_text += f"\n{'='*50}\n"
+                all_results_text += f"迭代 {i}\n"
+                all_results_text += f"{'='*50}\n"
+                
+                for r in results:
+                    all_results_text += f"\n[{r['timestamp']}] {r['phase']}\n"
+                    all_results_text += f"  {r['content']}\n"
+                    all_results_text += f"  {r['detail']}\n"
                 
                 # 更新图表数据
-                iterations_data.append({"iteration": i, "score": current_score})
+                chart_data.append({"iteration": i, "score": current_score})
                 
-                yield {
-                    "status": f"第 {i} 次迭代完成",
-                    "iteration": i,
-                    "score": current_score
-                }, all_results, gr.update(
-                    value=iterations_data,
-                    visible=True
-                )
+                status = f"状态: 第 {i} 次迭代完成 (分数: {current_score:.2f})"
+                yield status, pd.DataFrame(chart_data), all_results_text
                 
                 # 检查是否达到目标
                 if current_score >= target:
                     break
                 
-                time.sleep(0.5)
+                time.sleep(0.3)
             
-            final_status = {
-                "status": "任务完成" if current_score >= target else "达到最大迭代",
-                "final_score": current_score,
-                "iterations": len(iterations_data)
-            }
-            
-            yield final_status, all_results, gr.update(
-                value=iterations_data,
-                visible=True
-            )
+            final_status = f"状态: 任务完成\n最终分数: {current_score:.2f}\n总迭代次数: {len(chart_data)}"
+            yield final_status, pd.DataFrame(chart_data), all_results_text
         
         run_btn.click(
             fn=run_task,
             inputs=[task_input, max_iterations, target_score],
-            outputs=[status_output, iteration_output, score_chart]
+            outputs=[status_output, score_chart, iteration_output]
         )
         
         gr.Markdown("""
@@ -305,10 +289,10 @@ def create_demo():
         
         LoongFlow 是一个面向复杂任务的进化式 Agent 框架，特别适用于：
         
-        - 📐 **数学推理**: 开放式数学问题求解
-        - 🤖 **机器学习**: AutoML 和算法优化
-        - 💻 **代码生成**: 复杂编程任务
-        - 🔬 **科学研究**: 实验设计和分析
+        - **数学推理**: 开放式数学问题求解
+        - **机器学习**: AutoML 和算法优化
+        - **代码生成**: 复杂编程任务
+        - **科学研究**: 实验设计和分析
         
         了解更多: [GitHub](https://github.com/your-repo/loongflow)
         """)