Spaces:

ahaahaaha
/

adaptive_rag

Paused

App Files Files Community

lanny xu commited on Nov 11, 2025

Commit

e0f2417

1 Parent(s): 9b75bde

delete vectara

Browse files

Files changed (2) hide show

kaggle_flask_app.py +448 -0
kaggle_gradio_app.py +212 -0

kaggle_flask_app.py ADDED Viewed

	@@ -0,0 +1,448 @@

+"""
+Kaggle Flask 智能问答界面
+在 Kaggle Notebook 中使用 Flask 创建交互式 RAG 系统
+"""
+from flask import Flask, render_template_string, request, jsonify
+import sys
+import os
+import threading
+# 添加项目路径
+if '/kaggle/working/adaptive_RAG' not in sys.path:
+    sys.path.insert(0, '/kaggle/working/adaptive_RAG')
+from main import AdaptiveRAGSystem
+# 创建Flask应用
+app = Flask(__name__)
+# 全局RAG系统实例
+rag_system = None
+initialization_error = None
+# HTML模板
+HTML_TEMPLATE = """
+<!DOCTYPE html>
+<html lang="zh-CN">
+<head>
+    <meta charset="UTF-8">
+    <meta name="viewport" content="width=device-width, initial-scale=1.0">
+    <title>🤖 自适应RAG智能问答</title>
+    <style>
+        * {
+            margin: 0;
+            padding: 0;
+            box-sizing: border-box;
+        }
+        body {
+            font-family: 'Segoe UI', Tahoma, Geneva, Verdana, sans-serif;
+            background: linear-gradient(135deg, #667eea 0%, #764ba2 100%);
+            min-height: 100vh;
+            display: flex;
+            justify-content: center;
+            align-items: center;
+            padding: 20px;
+        }
+        .container {
+            background: white;
+            border-radius: 20px;
+            box-shadow: 0 20px 60px rgba(0,0,0,0.3);
+            width: 100%;
+            max-width: 800px;
+            overflow: hidden;
+        }
+        .header {
+            background: linear-gradient(135deg, #667eea 0%, #764ba2 100%);
+            color: white;
+            padding: 30px;
+            text-align: center;
+        }
+        .header h1 {
+            font-size: 2em;
+            margin-bottom: 10px;
+        }
+        .header p {
+            opacity: 0.9;
+            font-size: 0.9em;
+        }
+        .chat-container {
+            height: 500px;
+            overflow-y: auto;
+            padding: 20px;
+            background: #f7f7f7;
+        }
+        .message {
+            margin-bottom: 20px;
+            display: flex;
+            align-items: flex-start;
+        }
+        .message.user {
+            flex-direction: row-reverse;
+        }
+        .message-content {
+            max-width: 70%;
+            padding: 15px 20px;
+            border-radius: 18px;
+            position: relative;
+            word-wrap: break-word;
+        }
+        .message.user .message-content {
+            background: #667eea;
+            color: white;
+            margin-left: auto;
+        }
+        .message.bot .message-content {
+            background: white;
+            color: #333;
+            border: 1px solid #e0e0e0;
+        }
+        .avatar {
+            width: 40px;
+            height: 40px;
+            border-radius: 50%;
+            display: flex;
+            align-items: center;
+            justify-content: center;
+            font-size: 1.5em;
+            margin: 0 10px;
+        }
+        .input-container {
+            padding: 20px;
+            background: white;
+            border-top: 1px solid #e0e0e0;
+        }
+        .input-group {
+            display: flex;
+            gap: 10px;
+        }
+        #question-input {
+            flex: 1;
+            padding: 15px 20px;
+            border: 2px solid #e0e0e0;
+            border-radius: 25px;
+            font-size: 1em;
+            outline: none;
+            transition: border-color 0.3s;
+        }
+        #question-input:focus {
+            border-color: #667eea;
+        }
+        #send-btn {
+            padding: 15px 30px;
+            background: linear-gradient(135deg, #667eea 0%, #764ba2 100%);
+            color: white;
+            border: none;
+            border-radius: 25px;
+            cursor: pointer;
+            font-size: 1em;
+            font-weight: bold;
+            transition: transform 0.2s;
+        }
+        #send-btn:hover {
+            transform: scale(1.05);
+        }
+        #send-btn:disabled {
+            opacity: 0.5;
+            cursor: not-allowed;
+        }
+        .examples {
+            padding: 20px;
+            background: #f0f0f0;
+        }
+        .examples h3 {
+            margin-bottom: 10px;
+            color: #667eea;
+        }
+        .example-btn {
+            display: inline-block;
+            margin: 5px;
+            padding: 8px 15px;
+            background: white;
+            border: 1px solid #667eea;
+            color: #667eea;
+            border-radius: 15px;
+            cursor: pointer;
+            font-size: 0.9em;
+            transition: all 0.3s;
+        }
+        .example-btn:hover {
+            background: #667eea;
+            color: white;
+        }
+        .loading {
+            display: none;
+            text-align: center;
+            padding: 10px;
+            color: #667eea;
+        }
+        .error-message {
+            background: #fee;
+            color: #c33;
+            padding: 15px;
+            margin: 10px 0;
+            border-radius: 10px;
+            border-left: 4px solid #c33;
+        }
+        .status {
+            padding: 10px 20px;
+            text-align: center;
+            font-size: 0.9em;
+        }
+        .status.ok {
+            background: #d4edda;
+            color: #155724;
+        }
+        .status.error {
+            background: #f8d7da;
+            color: #721c24;
+        }
+    </style>
+</head>
+<body>
+    <div class="container">
+        <div class="header">
+            <h1>🤖 自适应RAG智能问答</h1>
+            <p>基于 LangGraph 的检索增强生成系统</p>
+        </div>
+        <div class="status {{ 'ok' if system_ok else 'error' }}">
+            {{ '✅ 系统运行正常' if system_ok else '❌ 系统初始化失败: ' + error }}
+        </div>
+        <div class="chat-container" id="chat-container">
+            <div class="message bot">
+                <div class="avatar">🤖</div>
+                <div class="message-content">
+                    你好！我是自适应RAG智能助手。我可以帮你回答关于LLM、Agent、Prompt Engineering等问题。请输入你的问题！
+                </div>
+            </div>
+        </div>
+        <div class="examples">
+            <h3>💡 示例问题</h3>
+            <button class="example-btn" onclick="askExample('AlphaCodium论文讲的是什么？')">
+                AlphaCodium论文讲的是什么？
+            </button>
+            <button class="example-btn" onclick="askExample('解释embedding嵌入的原理')">
+                解释embedding嵌入的原理
+            </button>
+            <button class="example-btn" onclick="askExample('什么是LLM Agent？')">
+                什么是LLM Agent？
+            </button>
+            <button class="example-btn" onclick="askExample('如何防止LLM产生幻觉？')">
+                如何防止LLM产生幻觉？
+            </button>
+        </div>
+        <div class="loading" id="loading">
+            ⏳ AI正在思考中...
+        </div>
+        <div class="input-container">
+            <div class="input-group">
+                <input
+                    type="text"
+                    id="question-input"
+                    placeholder="输入你的问题..."
+                    onkeypress="handleKeyPress(event)"
+                >
+                <button id="send-btn" onclick="sendQuestion()">🚀 发送</button>
+            </div>
+        </div>
+    </div>
+    <script>
+        function addMessage(content, isUser) {
+            const chatContainer = document.getElementById('chat-container');
+            const messageDiv = document.createElement('div');
+            messageDiv.className = 'message ' + (isUser ? 'user' : 'bot');
+            messageDiv.innerHTML = `
+                <div class="avatar">${isUser ? '👤' : '🤖'}</div>
+                <div class="message-content">${content}</div>
+            `;
+            chatContainer.appendChild(messageDiv);
+            chatContainer.scrollTop = chatContainer.scrollHeight;
+        }
+        function askExample(question) {
+            document.getElementById('question-input').value = question;
+            sendQuestion();
+        }
+        function handleKeyPress(event) {
+            if (event.key === 'Enter') {
+                sendQuestion();
+            }
+        }
+        async function sendQuestion() {
+            const input = document.getElementById('question-input');
+            const question = input.value.trim();
+            if (!question) {
+                return;
+            }
+            // 添加用户消息
+            addMessage(question, true);
+            input.value = '';
+            // 显示加载状态
+            const loading = document.getElementById('loading');
+            const sendBtn = document.getElementById('send-btn');
+            loading.style.display = 'block';
+            sendBtn.disabled = true;
+            try {
+                const response = await fetch('/api/query', {
+                    method: 'POST',
+                    headers: {
+                        'Content-Type': 'application/json',
+                    },
+                    body: JSON.stringify({ question: question })
+                });
+                const data = await response.json();
+                if (data.success) {
+                    addMessage(data.answer, false);
+                } else {
+                    addMessage('❌ 错误: ' + data.error, false);
+                }
+            } catch (error) {
+                addMessage('❌ 网络错误: ' + error.message, false);
+            } finally {
+                loading.style.display = 'none';
+                sendBtn.disabled = false;
+            }
+        }
+    </script>
+</body>
+</html>
+"""
+@app.route('/')
+def index():
+    """主页"""
+    return render_template_string(
+        HTML_TEMPLATE,
+        system_ok=(rag_system is not None),
+        error=initialization_error or ""
+    )
+@app.route('/api/query', methods=['POST'])
+def query():
+    """处理查询请求"""
+    if rag_system is None:
+        return jsonify({
+            'success': False,
+            'error': f'系统未初始化: {initialization_error}'
+        })
+    try:
+        data = request.get_json()
+        question = data.get('question', '').strip()
+        if not question:
+            return jsonify({
+                'success': False,
+                'error': '问题不能为空'
+            })
+        # 查询RAG系统
+        result = rag_system.query(question, verbose=False)
+        answer = result.get('answer', '无法生成答案')
+        return jsonify({
+            'success': True,
+            'answer': answer,
+            'metrics': result.get('retrieval_metrics')
+        })
+    except Exception as e:
+        return jsonify({
+            'success': False,
+            'error': str(e)
+        })
+def initialize_rag():
+    """初始化RAG系统"""
+    global rag_system, initialization_error
+    print("🚀 正在初始化RAG系统...")
+    try:
+        rag_system = AdaptiveRAGSystem()
+        print("✅ RAG系统初始化成功")
+    except Exception as e:
+        initialization_error = str(e)
+        print(f"❌ RAG系统初始化失败: {e}")
+def run_flask(host='0.0.0.0', port=5000):
+    """
+    启动Flask应用
+    Args:
+        host: 主机地址
+        port: 端口号
+    """
+    print("=" * 60)
+    print("🚀 启动 Flask RAG 智能问答系统")
+    print("=" * 60)
+    # 初始化RAG系统
+    initialize_rag()
+    # 检查是否在Kaggle环境
+    is_kaggle = os.path.exists('/kaggle/working')
+    if is_kaggle:
+        print("🎯 检测到 Kaggle 环境")
+        print(f"💡 访问地址: http://localhost:{port}")
+        print("⚠️  注意: Kaggle无法从外部访问，只能在Notebook内查看")
+    print(f"\n🌐 正在启动服务...")
+    print(f"   地址: http://{host}:{port}")
+    print("=" * 60)
+    # 启动Flask
+    app.run(host=host, port=port, debug=False, threaded=True)
+if __name__ == "__main__":
+    run_flask(port=5000)

kaggle_gradio_app.py ADDED Viewed

	@@ -0,0 +1,212 @@

+"""
+Kaggle Gradio 智能问答界面
+适合在 Kaggle Notebook 中运行的交互式 RAG 系统
+"""
+import gradio as gr
+import sys
+import os
+# 添加项目路径
+if '/kaggle/working/adaptive_RAG' not in sys.path:
+    sys.path.insert(0, '/kaggle/working/adaptive_RAG')
+from main import AdaptiveRAGSystem
+class RAGChatInterface:
+    """RAG聊天界面"""
+    def __init__(self):
+        """初始化RAG系统"""
+        print("🚀 正在初始化RAG系统...")
+        try:
+            self.rag_system = AdaptiveRAGSystem()
+            self.initialized = True
+            print("✅ RAG系统初始化成功")
+        except Exception as e:
+            print(f"❌ RAG系统初始化失败: {e}")
+            self.initialized = False
+            self.error_message = str(e)
+    def chat(self, message, history):
+        """
+        处理聊天消息
+        Args:
+            message: 用户输入的消息
+            history: 聊天历史记录
+        Returns:
+            响应消息
+        """
+        if not self.initialized:
+            return f"❌ 系统未初始化: {self.error_message}"
+        if not message or not message.strip():
+            return "⚠️ 请输入有效的问题"
+        try:
+            # 查询RAG系统
+            result = self.rag_system.query(message, verbose=False)
+            # 构建响应
+            answer = result.get('answer', '无法生成答案')
+            # 添加评估指标（可选）
+            metrics = result.get('retrieval_metrics')
+            if metrics:
+                metrics_text = f"\n\n📊 检索指标:\n"
+                metrics_text += f"- 耗时: {metrics.get('latency', 0):.2f}秒\n"
+                metrics_text += f"- 文档数: {metrics.get('retrieved_docs_count', 0)}\n"
+                metrics_text += f"- Precision@3: {metrics.get('precision_at_3', 0):.2f}\n"
+                # answer += metrics_text  # 取消注释以显示指标
+            return answer
+        except Exception as e:
+            return f"❌ 查询失败: {str(e)}"
+    def create_interface(self):
+        """创建Gradio界面"""
+        # 自定义CSS样式
+        custom_css = """
+        .gradio-container {
+            font-family: 'Arial', sans-serif;
+        }
+        .chatbot {
+            height: 500px;
+        }
+        """
+        # 创建聊天界面
+        with gr.Blocks(css=custom_css, title="🤖 自适应RAG智能问答") as demo:
+            gr.Markdown(
+                """
+                # 🤖 自适应RAG智能问答系统
+                基于LangGraph的自适应检索增强生成系统，支持：
+                - 🔍 智能路由（本地知识库 vs 网络搜索）
+                - 📚 混合检索（向量 + BM25）
+                - 🎯 多重质量控制（文档评分、幻觉检测）
+                - 🔄 自适应查询重写
+                **使用方法**: 在下方输入框输入问题，系统会自动选择最佳检索策略并生成答案。
+                """
+            )
+            # 聊天界面
+            chatbot = gr.Chatbot(
+                label="对话历史",
+                height=500,
+                show_label=True,
+                avatar_images=(None, "🤖")
+            )
+            with gr.Row():
+                msg = gr.Textbox(
+                    label="输入问题",
+                    placeholder="例如: AlphaCodium论文讲的是什么？",
+                    lines=2,
+                    scale=4
+                )
+                submit_btn = gr.Button("🚀 发送", scale=1, variant="primary")
+            with gr.Row():
+                clear_btn = gr.Button("🗑️ 清空对话", scale=1)
+            # 示例问题
+            gr.Examples(
+                examples=[
+                    "AlphaCodium论文讲的是什么？",
+                    "解释embedding嵌入的原理",
+                    "什么是LLM Agent？",
+                    "如何防止LLM产生幻觉？",
+                    "Prompt Engineering的最佳实践是什么？"
+                ],
+                inputs=msg,
+                label="💡 示例问题"
+            )
+            # 状态信息
+            if self.initialized:
+                gr.Markdown("✅ **系统状态**: 运行正常")
+            else:
+                gr.Markdown(f"❌ **系统状态**: 初始化失败 - {self.error_message}")
+            # 事件绑定
+            def respond(message, chat_history):
+                """响应用户消息"""
+                if not message:
+                    return "", chat_history
+                # 添加用户消息到历史
+                chat_history.append((message, None))
+                # 获取AI响���
+                bot_message = self.chat(message, chat_history)
+                # 更新历史
+                chat_history[-1] = (message, bot_message)
+                return "", chat_history
+            # 绑定发送按钮
+            submit_btn.click(
+                respond,
+                inputs=[msg, chatbot],
+                outputs=[msg, chatbot]
+            )
+            # 绑定回车键
+            msg.submit(
+                respond,
+                inputs=[msg, chatbot],
+                outputs=[msg, chatbot]
+            )
+            # 清空对话
+            clear_btn.click(lambda: None, None, chatbot, queue=False)
+        return demo
+def launch_app(share=False, server_port=7860):
+    """
+    启动Gradio应用
+    Args:
+        share: 是否创建公开链接（Kaggle中建议False）
+        server_port: 服务器端口
+    """
+    print("=" * 60)
+    print("🚀 启动 Gradio RAG 智能问答系统")
+    print("=" * 60)
+    # 检查是否在Kaggle环境
+    is_kaggle = os.path.exists('/kaggle/working')
+    if is_kaggle:
+        print("🎯 检测到 Kaggle 环境")
+        print("💡 提示: 运行后会显示本地URL")
+    # 创建界面
+    interface = RAGChatInterface()
+    demo = interface.create_interface()
+    # 启动服务
+    print(f"\n🌐 正在启动服务...")
+    demo.launch(
+        share=share,
+        server_port=server_port,
+        server_name="0.0.0.0",  # 允许所有IP访问
+        show_error=True,
+        quiet=False
+    )
+if __name__ == "__main__":
+    # 在Kaggle Notebook中运行时自动启动
+    launch_app(share=False)