add: rl-dpo

scripts/static/js/script.js

@@ -12,6 +12,14 @@ function openTab(evt, tabName) {
     document.getElementById(tabName).classList.remove("hidden");
     evt.currentTarget.classList.add("active");
     
+    // 当切换到其他标签页时，清除所有日志自动更新定时器
+    if (tabName !== 'processes') {
+        for (let processId in logTimers) {
+            clearInterval(logTimers[processId]);
+            delete logTimers[processId];
+        }
+    }
+    
     // 如果切换到进程页面，刷新进程列表
     if (tabName === 'processes') {
         loadProcesses();
@@ -25,6 +33,8 @@ document.getElementById('train_type').addEventListener('change', function() {
     const trainType = this.value;
     const pretrainSftFields = document.querySelectorAll('.pretrain-sft');
     const loraFields = document.querySelectorAll('.lora');
+    const dpoFields = document.querySelectorAll('.dpo');
+    const dpoParamCard = document.querySelector('.parameter-card.dpo');
     
     pretrainSftFields.forEach(field => {
         field.style.display = (trainType === 'pretrain' || trainType === 'sft') ? 'block' : 'none';
@@ -34,6 +44,14 @@ document.getElementById('train_type').addEventListener('change', function() {
         field.style.display = trainType === 'lora' ? 'block' : 'none';
     });
     
+    dpoFields.forEach(field => {
+        field.style.display = trainType === 'dpo' ? 'block' : 'none';
+    });
+    
+    if (dpoParamCard) {
+        dpoParamCard.style.display = trainType === 'dpo' ? 'block' : 'none';
+    }
+    
     // 设置默认值
     if (trainType === 'pretrain') {
         document.getElementById('save_dir').value = '../out';
@@ -65,6 +83,17 @@ document.getElementById('train_type').addEventListener('change', function() {
         document.getElementById('from_weight').value = 'full_sft';
         document.getElementById('log_interval').value = '10';
         document.getElementById('save_interval').value = '1';
+    } else if (trainType === 'dpo') {
+        document.getElementById('save_dir').value = '../out';
+        document.getElementById('save_weight').value = 'dpo';
+        document.getElementById('epochs').value = '1';
+        document.getElementById('batch_size').value = '4';
+        document.getElementById('learning_rate').value = '4e-8';
+        document.getElementById('data_path').value = '../dataset/dpo.jsonl';
+        document.getElementById('from_weight').value = 'full_sft';
+        document.getElementById('log_interval').value = '100';
+        document.getElementById('save_interval').value = '100';
+        document.getElementById('beta').value = '0.1';
     }
 });
 
@@ -100,6 +129,10 @@ function loadProcesses() {
                     statusClass = 'status-completed';
                 }
                 
+                // 设置进程数据属性，用于后续检查状态
+                processItem.dataset.processId = process.id;
+                processItem.dataset.processStatus = process.status;
+                
                 processItem.innerHTML = `
                     <div class="process-info">
                         <div>
@@ -122,13 +155,52 @@ function loadProcesses() {
         });
 }
 
+// 存储训练日志定时器的ID
+let logTimers = {};
+
 // 显示日志
 function showLogs(processId) {
     const logsContainer = document.getElementById(`logs-${processId}`);
     logsContainer.classList.toggle('hidden');
     
     if (!logsContainer.classList.contains('hidden')) {
+        // 加载日志内容
         loadLogContent(processId, logsContainer);
+        
+        // 清除可能存在的旧定时器
+        if (logTimers[processId]) {
+            clearInterval(logTimers[processId]);
+        }
+        
+        // 查找进程元素，获取其状态
+        const processItem = document.querySelector(`[data-process-id="${processId}"]`);
+        const isRunning = processItem && processItem.dataset.processStatus === '运行中';
+        
+        // 只有运行中的进程才设置自动刷新定时器
+        if (isRunning) {
+            logTimers[processId] = setInterval(() => {
+                // 检查日志容器是否可见
+                if (!logsContainer.classList.contains('hidden')) {
+                    // 再次检查进程状态，确保仍然是运行中
+                    const currentProcessItem = document.querySelector(`[data-process-id="${processId}"]`);
+                    const stillRunning = currentProcessItem && currentProcessItem.dataset.processStatus === '运行中';
+                    
+                    if (stillRunning) {
+                        loadLogContent(processId, logsContainer);
+                    } else {
+                        // 如果进程不再运行中，清除定时器
+                        clearInterval(logTimers[processId]);
+                        delete logTimers[processId];
+                    }
+                }
+            }, 1000);
+        }
+    } else {
+        // 隐藏日志时清除定时器
+        if (logTimers[processId]) {
+            clearInterval(logTimers[processId]);
+            delete logTimers[processId];
+        }
     }
 }
 
@@ -137,6 +209,34 @@ function refreshLog(processId) {
     const logsContainer = document.getElementById(`logs-${processId}`);
     if (!logsContainer.classList.contains('hidden')) {
         loadLogContent(processId, logsContainer);
+        
+        // 查找进程元素，获取其状态
+        const processItem = document.querySelector(`[data-process-id="${processId}"]`);
+        const isRunning = processItem && processItem.dataset.processStatus === '运行中';
+        
+        // 重置定时器，确保从现在开始每1秒更新一次
+        if (logTimers[processId]) {
+            clearInterval(logTimers[processId]);
+        }
+        
+        // 只有运行中的进程才设置定时器
+        if (isRunning) {
+            logTimers[processId] = setInterval(() => {
+                if (!logsContainer.classList.contains('hidden')) {
+                    // 再次检查进程状态
+                    const currentProcessItem = document.querySelector(`[data-process-id="${processId}"]`);
+                    const stillRunning = currentProcessItem && currentProcessItem.dataset.processStatus === '运行中';
+                    
+                    if (stillRunning) {
+                        loadLogContent(processId, logsContainer);
+                    } else {
+                        // 如果进程不再运行中，清除定时器
+                        clearInterval(logTimers[processId]);
+                        delete logTimers[processId];
+                    }
+                }
+            }, 1000);
+        }
     }
 }
 
@@ -214,7 +314,16 @@ function loadLogFiles() {
                     trainType = 'sft';
                 } else if (logfile.filename.includes('train_lora_')) {
                     trainType = 'lora';
+                } else if (logfile.filename.includes('train_dpo_')) {
+                    trainType = 'dpo';
+                } else if (logfile.filename.includes('train_ppo_')) {
+                    trainType = 'ppo';
+                } else if (logfile.filename.includes('train_grpo_')) {
+                    trainType = 'grpo';
+                } else if (logfile.filename.includes('train_spo_')) {
+                    trainType = 'spo';
                 }
+
                 
                 fileItem.innerHTML = `
                     <div class="process-info">

scripts/templates/index.html

@@ -3,11 +3,11 @@
 <head>
     <meta charset="UTF-8">
     <meta name="viewport" content="width=device-width, initial-scale=1.0">
-    <title>MiniMind 训练 Web UI</title>
+    <title>MiniMind Training Lab</title>
     <link rel="stylesheet" href="/static/css/style.css">
 </head>
 <body>
-    <h1>MiniMind 训练 Web UI</h1>
+    <h1>MiniMind Training Lab</h1>
     
     <div class="tabs">
         <button class="tab active" onclick="openTab(event, 'train')">开始训练</button>
@@ -29,6 +29,7 @@
                                 <option value="pretrain">预训练 (Pretrain)</option>
                                 <option value="sft">监督微调 (Full SFT)</option>
                                 <option value="lora">LoRA 微调</option>
+                                <option value="dpo">强化学习 (RL-DPO)</option>
                             </select>
                         </div>
                         <div class="form-group">
@@ -54,6 +55,17 @@
                     </div>
                 </div>
                 
+                <!-- 强化学习参数 -->
+                <div class="parameter-card dpo" style="display: none;">
+                    <h3 class="card-title">强化学习参数</h3>
+                    <div class="parameter-content">
+                        <div class="form-group">
+                            <label for="beta">DPO Beta 参数：</label>
+                            <input type="text" id="beta" name="beta" placeholder="0.1">
+                        </div>
+                    </div>
+                </div>
+                
                 <!-- 模型结构参数 -->
                 <div class="parameter-card">
                     <h3 class="card-title">模型结构参数</h3>

scripts/train_web_ui.py

@@ -43,13 +43,23 @@ def start_training_process(train_type, params):
         cmd = [sys.executable, script_path]
         if 'lora_name' in params:
             cmd.extend(['--lora_name', params['lora_name']])
+    elif train_type == 'dpo':
+        script_path = '../trainer/train_dpo.py'
+        cmd = [sys.executable, script_path]
+        # 添加DPO特定参数
+        if 'beta' in params and params['beta']:
+            cmd.extend(['--beta', params['beta']])
+        if 'accumulation_steps' in params and params['accumulation_steps']:
+            cmd.extend(['--accumulation_steps', params['accumulation_steps']])
+        if 'grad_clip' in params and params['grad_clip']:
+            cmd.extend(['--grad_clip', params['grad_clip']])
     else:
         return None
     
     # 添加通用参数
     for key, value in params.items():
-        # 跳过特殊参数
+        # 跳过特殊参数和DPO特有参数
-        if key in ['train_type', 'save_weight', 'lora_name', 'train_monitor']:
+        if key in ['train_type', 'save_weight', 'lora_name', 'train_monitor', 'beta', 'accumulation_steps', 'grad_clip']:
             continue
             
         # 特殊处理布尔标志参数
@@ -64,6 +74,8 @@ def start_training_process(train_type, params):
     if 'train_monitor' in params:
         if params['train_monitor'] == 'wandb' or params['train_monitor'] == 'swanlab':
             cmd.append('--use_wandb')  # 对于wandb和swanlab，只添加标志，不添加值
+            if params['train_monitor'] == 'wandb':
+                cmd.extend(['--wandb_project', 'minimind_training'])
     
     # 创建日志文件
     with open(log_file, 'w') as f:
@@ -170,10 +182,10 @@ def logs(process_id):
                 if os.path.exists(log_file):
                     try:
                         with open(log_file, 'r', encoding='utf-8') as f:
-                            # 只读取最后500行
+                            # 只读取最后200行
                             lines = f.readlines()
-                            last_500_lines = lines[-500:] if len(lines) > 500 else lines
+                            last_200_lines = lines[-200:] if len(lines) > 200 else lines
-                            return ''.join(last_500_lines)
+                            return ''.join(last_200_lines)
                     except Exception as e:
                         return f'读取日志失败: {str(e)}'
     return '日志文件不存在或已被删除'