feat: 关于页加版本 v2.0 chip + 35 因子表（标缺失项）

Author: Johnny-xuan

docs/index.html

@@ -57,6 +57,7 @@
   <a class="brand" href="#picks" data-tab-link="picks">
     <span class="brand__mark">M<sup>2</sup></span>
     <span class="brand__name">α<em>lpha</em></span>
+    <span class="brand__ver">v2.0</span>
   </a>
 
   <ul class="nav-tabs">
@@ -398,6 +399,71 @@ <h2 class="about-h">M²·Alpha 是什么</h2>
         </p>
       </article>
 
+      <article class="about-block">
+        <h2 class="about-h">模型版本 · <span class="ver-chip">v2.0</span></h2>
+        <p>
+          当前线上是 <strong>M²-Alpha v2.0</strong>，基于 SeqMASTER 架构 + 35 个 basic features，
+          沪深 300 截面排名任务、CPU 推理（~600K 参数 / 2.4 MB ckpt）。
+        </p>
+        <ul class="ver-roadmap">
+          <li><span class="ver-tag ver-tag--cur">v2.0 现行</span>basic 35 因子 · BaoStock 数据源 · 行业分散 + 滞后带策略</li>
+          <li><span class="ver-tag ver-tag--next">v2.1 计划</span>切 Tushare 数据源，补齐目前缺失的 3 个 derived features</li>
+          <li><span class="ver-tag ver-tag--next">v3.0 计划</span>引入 market gating + cross-stock attention，增强宏观敏感性</li>
+          <li><span class="ver-tag ver-tag--next">v4.0 设想</span>多模型 ensemble + 自适应换手带宽</li>
+        </ul>
+        <p class="dim">
+          更高版本敬请期待，每次升级会带 changelog 与新旧 OOS 对比一并发布。
+        </p>
+      </article>
+
+      <article class="about-block">
+        <h2 class="about-h">35 个输入因子（v2.0 全公开）</h2>
+        <p class="dim" style="margin-top:0">
+          所有因子按个股时序计算，再做截面 robust z-score 后喂模型。
+          带 <span class="feat-warn">⚠</span> 的因子目前用 BaoStock 数据无法填实数（恒为 0，等价于中性信号），
+          v2.1 切 Tushare 后可补齐。
+        </p>
+        <div class="feat-grid">
+          <div class="feat-group">
+            <h4 class="feat-h">动量 <span class="feat-cnt">8</span></h4>
+            <code class="feat-list">f_ret_1 · f_ret_2 · f_ret_3 · f_ret_5 · f_ret_10 · f_ret_20 · f_ret_30 · f_ret_60</code>
+          </div>
+          <div class="feat-group">
+            <h4 class="feat-h">均线偏离 <span class="feat-cnt">5</span></h4>
+            <code class="feat-list">f_close_ma_5 · f_close_ma_10 · f_close_ma_20 · f_close_ma_30 · f_close_ma_60</code>
+          </div>
+          <div class="feat-group">
+            <h4 class="feat-h">波动率 <span class="feat-cnt">5</span></h4>
+            <code class="feat-list">f_vol_5 · f_vol_10 · f_vol_20 · f_vol_30 · f_vol_60</code>
+          </div>
+          <div class="feat-group">
+            <h4 class="feat-h">K 线形态 <span class="feat-cnt">4</span></h4>
+            <code class="feat-list">f_kmid · f_klen · f_kup · f_klow</code>
+          </div>
+          <div class="feat-group">
+            <h4 class="feat-h">成交量 <span class="feat-cnt">5</span></h4>
+            <code class="feat-list">f_vol_ratio_5 · f_vol_ratio_10 · f_vol_ratio_20 · f_vol_chg_1 · f_amount_chg_1</code>
+          </div>
+          <div class="feat-group">
+            <h4 class="feat-h">换手 <span class="feat-cnt">2</span></h4>
+            <code class="feat-list">f_turnover_rate · f_turnover_rate_f</code>
+          </div>
+          <div class="feat-group">
+            <h4 class="feat-h">估值 <span class="feat-cnt">3</span></h4>
+            <code class="feat-list">f_pe_ttm · f_pb · f_ps_ttm</code>
+          </div>
+          <div class="feat-group feat-group--warn">
+            <h4 class="feat-h">缺失 / 数据源限制 <span class="feat-cnt feat-cnt--warn">3 <span class="feat-warn">⚠</span></span></h4>
+            <code class="feat-list">
+              f_volume_ratio <span class="dim">(量比)</span> ·
+              f_mf_net_ratio <span class="dim">(资金净流入率)</span> ·
+              f_mf_big_small_diff <span class="dim">(大单-小单差)</span>
+            </code>
+            <p class="feat-note">BaoStock 不暴露这 3 个字段，当前用 0 占位 → 模型实际只有 32 个有效信号。</p>
+          </div>
+        </div>
+      </article>
+
       <article class="about-block">
         <h2 class="about-h">交易策略 · 6 条规则</h2>
         <ol class="rules-compact">

docs/styles.css

@@ -154,6 +154,21 @@ body.light .navbar {
   font-variation-settings: "opsz" 60, "SOFT" 80, "WONK" 1;
 }
 .brand__name em { font-style: italic; color: var(--ink); font-weight: 400; }
+.brand__ver {
+  display: inline-block;
+  margin-left: 6px;
+  padding: 1px 6px;
+  font-family: var(--f-mono);
+  font-size: 10px;
+  font-weight: 500;
+  letter-spacing: 0.04em;
+  color: var(--ink-dim);
+  border: 1px solid var(--hairline-2);
+  border-radius: 3px;
+  vertical-align: middle;
+  background: rgba(255,255,255,0.03);
+}
+body.light .brand__ver { background: rgba(0,0,0,0.04); }
 
 .nav-tabs { list-style: none; display: flex; gap: 0; justify-content: center; }
 .nav-tab {
@@ -982,6 +997,112 @@ body.light .tooltip { box-shadow: 0 4px 18px rgba(0,0,0,0.12), 0 0 12px var(--li
 }
 .about-block ul li strong { color: var(--ink); }
 
+/* version chip in 模型版本 block title */
+.ver-chip {
+  display: inline-block;
+  padding: 2px 8px;
+  font-family: var(--f-mono);
+  font-size: 12px;
+  letter-spacing: 0.04em;
+  color: var(--lime);
+  border: 1px solid var(--lime);
+  border-radius: 4px;
+  background: rgba(200, 249, 61, 0.08);
+}
+body.light .ver-chip { background: rgba(140, 180, 0, 0.08); }
+
+/* roadmap list */
+.ver-roadmap {
+  list-style: none; padding: 0; margin: 12px 0 10px;
+  display: flex; flex-direction: column; gap: 6px;
+}
+.ver-roadmap li {
+  display: flex; align-items: baseline; gap: 12px;
+  padding: 6px 10px;
+  font-size: 12.5px; line-height: 1.5; color: var(--ink-dim);
+  border-radius: 4px;
+  background: var(--bg-2);
+}
+.ver-tag {
+  flex: 0 0 auto;
+  font-family: var(--f-mono); font-size: 10.5px;
+  padding: 2px 8px; border-radius: 3px;
+  letter-spacing: 0.03em; font-weight: 500;
+}
+.ver-tag--cur {
+  color: var(--bg);
+  background: var(--lime);
+}
+body.light .ver-tag--cur { color: var(--ink); }
+.ver-tag--next {
+  color: var(--ink-mute);
+  border: 1px solid var(--hairline);
+  background: transparent;
+}
+
+/* 35 features grid */
+.feat-grid {
+  display: grid;
+  grid-template-columns: repeat(2, minmax(0, 1fr));
+  gap: 8px;
+  margin-top: 12px;
+}
+.feat-group {
+  padding: 10px 12px;
+  border: 1px solid var(--hairline);
+  border-radius: 4px;
+  background: var(--bg-2);
+}
+.feat-group--warn {
+  grid-column: 1 / -1;
+  border-color: var(--amber);
+  background: rgba(212, 160, 23, 0.06);
+}
+body.light .feat-group--warn { background: rgba(180, 130, 0, 0.05); }
+.feat-h {
+  margin: 0 0 6px;
+  font-family: var(--f-sans);
+  font-size: 12px;
+  letter-spacing: 0.03em;
+  color: var(--ink);
+  font-weight: 600;
+  display: flex; align-items: center; gap: 8px;
+}
+.feat-cnt {
+  font-family: var(--f-mono); font-size: 10px;
+  padding: 1px 6px; border-radius: 2px;
+  color: var(--ink-mute);
+  background: rgba(255,255,255,0.04);
+  border: 1px solid var(--hairline);
+  font-weight: 400;
+}
+body.light .feat-cnt { background: rgba(0,0,0,0.03); }
+.feat-cnt--warn {
+  color: var(--amber);
+  border-color: var(--amber);
+  background: rgba(212, 160, 23, 0.08);
+}
+.feat-warn { color: var(--amber); font-weight: 700; }
+.feat-list {
+  display: block;
+  font-family: var(--f-mono);
+  font-size: 11px;
+  line-height: 1.7;
+  color: var(--ink-dim);
+  word-break: break-word;
+}
+.feat-list .dim { color: var(--ink-mute); font-size: 10.5px; }
+.feat-note {
+  margin: 8px 0 0;
+  font-size: 11.5px;
+  color: var(--ink-mute);
+  line-height: 1.55;
+}
+
+@media (max-width: 720px) {
+  .feat-grid { grid-template-columns: 1fr; }
+}
+
 .footer--about {
   margin-top: 32px;
   padding: 18px 0;