Hugging Face's logo

Spaces:
W1nd5pac
/
microclimate-x
Paused

App Files Community
microclimate-x / docs /progress_update_brief.html
W1nd5pac's picture
W1nd5pac
Deploy 2026-05-20T06:52:08Z — 11e81c5
4eefabb verified
raw
history blame contribute delete
36.3 kB
<!doctype html>
<html lang="en">
<head>
<meta charset="utf-8">
<title>Progress-Update Brief — MicroClimate-X</title>
<style>
 /* ============================================================
 Print-optimised A4 progress-update brief
 Open in browser → ⌘+P → Save as PDF, or read on screen
 ============================================================ */
 :root {
 --ink: #0b0d12;
 --ink-soft: #353a44;
 --muted: #6b7280;
 --brand: #2563eb;
 --brand-soft: #dbeafe;
 --accent: #b91c1c;
 --accent-soft: #fee2e2;
 --ok: #166534;
 --ok-soft: #dcfce7;
 --warn: #b45309;
 --warn-soft: #fef3c7;
 --grid: #e5e7eb;
 --bg: #ffffff;
 --code-bg: #f3f4f6;
 }
 * { box-sizing: border-box; }
 html, body { margin: 0; padding: 0; background: var(--bg); color: var(--ink); }
 body {
 font-family: -apple-system, BlinkMacSystemFont, "SF Pro Text",
 "PingFang SC", "Hiragino Sans GB", "Microsoft YaHei",
 system-ui, sans-serif;
 font-size: 11pt;
 line-height: 1.45;
 }
 @page { size: A4; margin: 12mm 14mm; }
 main { max-width: 200mm; margin: 0 auto; padding: 14mm 14mm; }
 /* Headings */
 h1 {
 font-size: 22pt; margin: 0 0 4mm 0;
 border-bottom: 3px solid var(--brand); padding-bottom: 3mm;
 page-break-after: avoid;
 }
 h1 .zh { display: block; font-size: 13pt; color: var(--muted); font-weight: 500; margin-top: 1mm; }
 h2 {
 font-size: 14pt; margin: 9mm 0 3mm 0;
 color: var(--brand);
 border-left: 4px solid var(--brand); padding: 1mm 0 1mm 3mm;
 page-break-after: avoid;
 }
 h2 .zh { display: block; font-size: 10pt; color: var(--muted); margin-top: 0.5mm; font-weight: 500; }
 h3 {
 font-size: 11.5pt; margin: 5mm 0 2mm 0; color: var(--ink-soft);
 page-break-after: avoid;
 }
 h4 { font-size: 10.5pt; margin: 3mm 0 1mm 0; color: var(--accent); }
 p, li { margin: 1mm 0; }
 ul, ol { padding-left: 5mm; }
 ul li { margin-bottom: 1mm; }
 /* Quote / supervisor verbatim */
 .quote {
 background: var(--warn-soft);
 border-left: 3px solid var(--warn);
 padding: 2mm 3mm; margin: 2mm 0;
 font-style: italic; font-size: 10pt;
 }
 .quote::before { content: "🎙️ "; font-style: normal; }
 /* Tables */
 table.bilingual, table.steps, table.tabs, table.plan, table {
 border-collapse: collapse; width: 100%; margin: 2mm 0 3mm 0;
 font-size: 10pt;
 }
 table.bilingual td, table.steps td, table.tabs td, table.plan td,
 table th, table td {
 padding: 1.5mm 2.5mm; vertical-align: top;
 border: 1px solid var(--grid);
 }
 table th {
 background: #f9fafb; font-weight: 600; text-align: left;
 color: var(--ink-soft);
 }
 table.bilingual td.en { width: 50%; }
 table.bilingual td.zh { width: 50%; background: #fafbfc; }
 table.plan td.estimate { width: 14%; text-align: center; color: var(--brand); font-weight: 600; }
 /* Inline callouts */
 .callout {
 margin: 2mm 0; padding: 2mm 3mm;
 border-left: 3px solid; border-radius: 1mm;
 font-size: 10pt;
 }
 .callout.warn { background: var(--accent-soft); border-color: var(--accent); }
 .callout.ok { background: var(--ok-soft); border-color: var(--ok); }
 .callout.tip { background: var(--brand-soft); border-color: var(--brand); }
 .callout-title { font-weight: 700; margin-bottom: 1mm; }
 /* Code */
 code, pre, kbd {
 font-family: "SF Mono", "JetBrains Mono", Menlo, Consolas, monospace;
 font-size: 9.5pt;
 }
 code { background: var(--code-bg); padding: 0.3mm 1mm; border-radius: 1mm; }
 pre {
 background: var(--code-bg); padding: 3mm; border-radius: 2mm;
 overflow-x: auto; margin: 2mm 0;
 border: 1px solid var(--grid);
 }
 pre code { background: transparent; padding: 0; }
 /* Step indicators */
 .step {
 display: flex; gap: 3mm;
 margin: 2mm 0;
 align-items: flex-start;
 }
 .step .num {
 flex: 0 0 8mm; width: 8mm; height: 8mm; border-radius: 50%;
 background: var(--brand); color: white; font-weight: 700;
 display: flex; align-items: center; justify-content: center;
 font-size: 11pt;
 }
 .step .body { flex: 1; }
 /* Demo / decision blocks */
 .demo {
 background: #f0f9ff;
 border: 1px solid #bae6fd;
 border-radius: 2mm;
 padding: 3mm;
 margin: 3mm 0;
 }
 .demo .demo-title { font-weight: 700; color: #075985; margin-bottom: 1mm; }
 .decision {
 background: #fefce8;
 border: 1px solid #fde047;
 border-radius: 2mm;
 padding: 3mm;
 margin: 2mm 0;
 }
 .decision .decision-title {
 font-weight: 700; color: #854d0e; margin-bottom: 1mm;
 text-transform: uppercase; letter-spacing: 0.5pt; font-size: 9pt;
 }
 /* Status pill */
 .pill {
 display: inline-block; padding: 0.2mm 1.5mm;
 border-radius: 4mm; font-size: 8.5pt; font-weight: 600;
 vertical-align: middle;
 }
 .pill.done { background: var(--ok-soft); color: var(--ok); }
 .pill.plan { background: var(--brand-soft); color: var(--brand); }
 .pill.risk { background: var(--accent-soft); color: var(--accent); }
 /* Checklist */
 .check { font-family: "SF Mono", Menlo, monospace; font-size: 9.5pt; line-height: 1.7; }
 .check .box { display: inline-block; width: 4mm; }
 /* Page break helpers */
 .pb { page-break-before: always; }
 .nobreak { page-break-inside: avoid; }
 /* Footer */
 footer {
 margin-top: 12mm; padding-top: 4mm;
 border-top: 1px solid var(--grid);
 color: var(--muted); font-size: 9pt; text-align: center;
 }
 /* Print refinements */
 @media print {
 body { font-size: 10pt; }
 h2 { font-size: 13pt; }
 .no-print { display: none; }
 a { color: var(--ink); text-decoration: none; }
 }
 /* Toolbar (screen only) */
 .toolbar {
 position: sticky; top: 0; z-index: 100;
 background: var(--brand); color: white;
 padding: 2mm 4mm; display: flex; justify-content: space-between;
 align-items: center; font-size: 10pt;
 }
 .toolbar button {
 background: white; color: var(--brand); border: 0;
 padding: 1.5mm 4mm; border-radius: 1mm; font-weight: 600;
 cursor: pointer; font-size: 10pt;
 }
 .toolbar button:hover { background: #f3f4f6; }
 /* Cover meta strip */
 .cover-meta {
 display: flex; gap: 4mm; flex-wrap: wrap;
 margin: 3mm 0;
 color: var(--muted); font-size: 9.5pt;
 }
 .cover-meta span {
 background: var(--code-bg); padding: 0.5mm 2mm; border-radius: 1mm;
 }
 /* Timeline strip in §0.2 */
 table.timeline td.block { width: 8%; text-align: center; font-weight: 700; color: var(--brand); }
 table.timeline td.time { width: 18%; font-family: "SF Mono", Menlo, monospace; color: var(--muted); }
</style>
</head>
<body>
<div class="toolbar no-print">
 <strong>Progress-Update Brief · MicroClimate-X</strong>
 <button onclick="window.print()">🖨 Print / Save as PDF</button>
</div>
<main>
<h1>Supervisor Progress-Update Brief
 <span class="zh">导师进度汇报双语逐字稿 — MicroClimate-X</span>
</h1>
<div class="cover-meta">
 <span>📅 2026-05-13</span>
 <span>🎓 UKM FYP</span>
 <span>🏛️ KyoukoLi/microclimate-x</span>
 <span>🚀 v1.0.0 shipped 2026-05-11</span>
 <span>✅ 70 tests · 97% coverage</span>
</div>
<div class="callout tip">
 <div class="callout-title">How to use this brief · 怎么用这份汇报稿</div>
 Follow-up meeting after the v1.0.0 hardening pass on 2026-05-11. Walk-through order is unchanged: <strong>dataset → model → app → next steps</strong>. Open this file on screen during the meeting; <strong>do not read word-for-word</strong>.<br><br>
 紧接 2026-05-11 v1.0.0 强化提交之后的<strong>进度汇报</strong>会议。顺序一律不变：<strong>dataset → model → app → 下一步</strong>。开会时屏幕上打开本文档，<strong>不要照念</strong>，当兜底用即可。
</div>
<!-- ===== Section 0: what you need to do ===== -->
<h2>0 · What you need to do — three time windows
 <span class="zh">你要做的事 —— 三个时间窗口</span>
</h2>
<h3>0.1 · Before the meeting (T-15 min) / 会前 15 分钟</h3>
<table class="check">
 <tr><th style="width:6%">☐</th><th>English</th><th>中文</th></tr>
 <tr><td>☐</td><td>Charge laptop ≥ 80 %; charger in bag.</td><td>笔记本充满 ≥ 80%，充电器带上。</td></tr>
 <tr><td>☐</td><td><code>cd ~/Projects/microclimate-x &amp;&amp; git pull &amp;&amp; git status</code> — must print "working tree clean".</td><td>拉最新代码，确认 working tree clean。</td></tr>
 <tr><td>☐</td><td><code>make run</code> in <strong>terminal A</strong> (leave it running).</td><td>终端 A 起后端，<strong>不要关</strong>。</td></tr>
 <tr><td>☐</td><td><code>curl -s http://localhost:8000/api/health | python3 -m json.tool</code> in <strong>terminal B</strong> — verify <code>"ml_loaded": true</code>.</td><td>终端 B 验证健康检查，<code>ml_loaded</code> 必须为 <code>true</code>。</td></tr>
 <tr><td>☐</td><td>Open the 10 browser tabs in the order from <code>MEETING_CHEAT_SHEET.md</code> §0 — <strong>app tab is last</strong>.</td><td>按 cheat-sheet §0 顺序开 10 个标签页，<strong>app 标签放最后</strong>。</td></tr>
 <tr><td>☐</td><td>This file (<code>progress_update_brief.html</code>) open on a separate screen / phone.</td><td>把本文档单独开在副屏或手机上。</td></tr>
 <tr><td>☐</td><td>Phone on silent. Deep breath.</td><td>手机静音，深呼吸。</td></tr>
</table>
<h3>0.2 · During the meeting (≈ 8 minutes) / 会中 ≈ 8 分钟</h3>
<table class="timeline">
 <tr><th>Block</th><th>EN heading</th><th>中文标题</th><th>Time</th></tr>
 <tr><td class="block">1</td><td>Opening 30 s</td><td>开场 30 秒</td><td class="time">0:00 → 0:30</td></tr>
 <tr><td class="block">2</td><td>What changed since last meeting</td><td>自上次会以来的进展</td><td class="time">0:30 → 2:00</td></tr>
 <tr><td class="block">3</td><td>Live demo — dataset → model → app</td><td>现场演示（顺序不变）</td><td class="time">2:00 → 5:00</td></tr>
 <tr><td class="block">4</td><td>Next steps for Chapter 5</td><td>Chapter 5 下一步</td><td class="time">5:00 → 6:30</td></tr>
 <tr><td class="block">5</td><td>Asks + closing</td><td>请示 + 收尾</td><td class="time">6:30 → 8:00</td></tr>
</table>
<h3>0.3 · After the meeting (T+24 h) / 会后 24 小时内</h3>
<table class="check">
 <tr><th style="width:6%">☐</th><th>English</th><th>中文</th></tr>
 <tr><td>☐</td><td>Write meeting minutes — capture every supervisor decision in <code>docs/meeting_log_&lt;date&gt;.md</code>.</td><td>写会议纪要，把老师每条决定记到 <code>docs/meeting_log_&lt;日期&gt;.md</code>。</td></tr>
 <tr><td>☐</td><td>Open one GitHub issue per agreed action item (label: <code>chapter-5</code>).</td><td>每个 action item 在 GitHub 开一个 issue，打 <code>chapter-5</code> 标签。</td></tr>
 <tr><td>☐</td><td>Email a 3-bullet summary back to the supervisor for written confirmation.</td><td>给老师发 3 条要点的总结邮件，留<strong>书面确认</strong>。</td></tr>
 <tr><td>☐</td><td>Update <code>README.md</code> §9 Roadmap — tick boxes that were signed off.</td><td>更新 <code>README.md</code> 第 9 节 Roadmap，把通过的项打勾。</td></tr>
 <tr><td>☐</td><td>Tag a new release if scope was confirmed (<code>git tag v1.1.0-rc.1</code>).</td><td>如果范围确认了，打个新 tag (<code>v1.1.0-rc.1</code>)。</td></tr>
</table>
<!-- ===== Section 1: Opening ===== -->
<h2 class="pb">1 · Opening (30 seconds)
 <span class="zh">开场 30 秒</span>
</h2>
<table class="bilingual">
 <tr>
 <td class="en">"Sir, thank you for your time. Following up on our last session, I've completed a production-grade hardening pass — version 1.0.0 — and the full pipeline is now reproducible end-to-end. May I walk you through what's new in the same order as before — <strong>dataset, then model, then app</strong> — and finish with my proposed plan for Chapter 5?"</td>
 <td class="zh">"老师感谢您抽时间。接着上次的内容，我做完了 <strong>v1.0.0 工程化强化</strong>，整条流水线现在可以<strong>端到端复现</strong>。我按上次的顺序——<strong>dataset、model、app</strong>——给您过一遍新的进展，最后讲我对 Chapter 5 的下一步计划，可以吗？"</td>
 </tr>
</table>
<div class="callout ok">
 <strong>Why this opening · 为什么这样开场</strong>: (a) restates the supervisor's preferred process order without him asking, (b) signals you've made forward progress (not just polish), (c) ends with an explicit ask for direction on Chapter 5 — which is what <em>he</em> wants to talk about.<br>
 (a) 不用他提就主动按他的流程顺序；(b) 强调是<strong>前进了</strong>而不是只在抛光；(c) 用对 Chapter 5 的请示收尾，<strong>这正是他想聊的话题</strong>。
</div>
<!-- ===== Section 2: progress recap ===== -->
<h2>2 · What changed since the last meeting
 <span class="zh">自上次会议以来的进展</span>
</h2>
<p style="color: var(--muted); font-size: 9.5pt;">
 ≈ 90 seconds. Stay on the GitHub repo tab — point to the commit history, the green CI badge, the v1.0.0 release.<br>
 ≈ 90 秒。停在 GitHub repo 标签页，指给老师看 commit 历史、CI 绿勾、v1.0.0 release。
</p>
<table class="bilingual">
 <tr><th style="width:18%">Area / 模块</th><th>English</th><th>中文</th></tr>
 <tr>
 <td><strong>Backend hardening</strong><br><span style="color:var(--muted);font-size:9pt">后端强化</span></td>
 <td>"I added a request-ID middleware, a typed <code>ErrorResponse</code> contract so no bare HTML 500s leak, structured logging, and an enriched <code>/api/health</code> exposing uptime, cache stats, and the loaded ML feature schema."</td>
 <td>"后端我加了 <strong>request-ID 中间件</strong>、<strong>类型化错误协议</strong> <code>ErrorResponse</code>（不再泄漏裸 HTML 500）、结构化日志、以及<strong>升级版 <code>/api/health</code></strong>（暴露 uptime、缓存统计、ML 特征 schema）。"</td>
 </tr>
 <tr>
 <td><strong>ML pipeline</strong><br><span style="color:var(--muted);font-size:9pt">ML 流水线</span></td>
 <td>"I shipped <code>scripts/4_evaluate_model.py</code> which produces six publication-quality figures plus a machine-readable <code>evaluation_summary.json</code>. I also wrote a HuggingFace-style <code>MODEL_CARD.md</code> covering intended use, training data, metrics, limitations, and ethical considerations."</td>
 <td>"ML 流水线加了<strong>评估脚本</strong> <code>scripts/4_evaluate_model.py</code>，自动出 6 张论文级别图 + 一份 <code>evaluation_summary.json</code>。还写了 HuggingFace 风格的 <strong>MODEL_CARD.md</strong>，覆盖用途、训练数据、指标、局限、伦理考量。"</td>
 </tr>
 <tr>
 <td><strong>Tests + CI</strong><br><span style="color:var(--muted);font-size:9pt">测试 + CI</span></td>
 <td>"Total tests went from 19 to <strong>70</strong>, backend coverage is <strong>97 %</strong>. CI runs on Python 3.9 / 3.11 / 3.12 plus a Docker image-build smoke test."</td>
 <td>"测试数从 19 涨到 <strong>70</strong>，<strong>后端覆盖率 97%</strong>。CI 跑 Python 3.9/3.11/3.12 矩阵，外加 Docker 镜像构建烟测。"</td>
 </tr>
 <tr>
 <td><strong>Dev-ex</strong><br><span style="color:var(--muted);font-size:9pt">开发体验</span></td>
 <td>"Multi-stage Dockerfile, docker-compose, Makefile single-word recipes, pre-commit hooks. The whole project is now <code>docker compose up --build</code> away from a clean machine."</td>
 <td>"多阶段 Dockerfile + compose + Makefile 单词命令 + pre-commit hooks。<strong>新机器一句 <code>docker compose up --build</code> 就能跑起来</strong>。"</td>
 </tr>
 <tr>
 <td><strong>Documentation</strong><br><span style="color:var(--muted);font-size:9pt">文档</span></td>
 <td>"Three new docs — <code>architecture.md</code>, <code>thresholds.md</code> with citations for every Veto threshold, and <code>pipeline_order.md</code> which explicitly enforces the dataset → model → app order you asked for."</td>
 <td>"三份新文档——<code>architecture.md</code>、<code>thresholds.md</code>（每个 Veto 阈值都附学术引用）、以及 <code>pipeline_order.md</code>（<strong>显式按您要求的 dataset→model→app 顺序写死</strong>）。"</td>
 </tr>
</table>
<div class="callout tip">
 <strong>Artefact to show · 展示物</strong>: GitHub commit history page; the green CI badge on the README; <code>CHANGELOG.md</code> v1.0.0 entry.<br>
 GitHub commit 历史页；README 上的 CI 绿勾；<code>CHANGELOG.md</code> 中 v1.0.0 那一段。
</div>
<!-- ===== Section 3: live demo ===== -->
<h2 class="pb">3 · Live demo — dataset → model → app
 <span class="zh">现场演示（顺序不变）</span>
</h2>
<p style="color: var(--muted); font-size: 9.5pt;">
 ≈ 3 minutes. Same order as the 5/11 dry-run script — no surprises for the supervisor.<br>
 ≈ 3 分钟。跟 5/11 的脚本完全一样的顺序，<strong>老师不会被打乱节奏</strong>。
</p>
<div class="step">
 <div class="num">1</div>
 <div class="body">
 <h3>Dataset (Tab <code>docs/dataset.md</code>) — 30 s</h3>
 <table class="bilingual">
 <tr>
 <td class="en">"Same dataset as last time — ERA5 reanalysis, 5 Malaysian mountain sites, 175 315 hourly rows. The Y column <code>is_rain_event</code> is derived in one line and documented in §5. <strong>No change here</strong>, just confirming the foundation is unchanged."</td>
 <td class="zh">"数据集跟上次一样——ERA5 再分析、马来西亚 5 个山地点位、17.5 万行小时数据。Y 列 <code>is_rain_event</code> 一行代码构造，文档在 §5。<strong>这里没有变</strong>，只是确认地基没动。"</td>
 </tr>
 </table>
 </div>
</div>
<div class="step">
 <div class="num">2</div>
 <div class="body">
 <h3>Model (Tabs <code>01_roc</code> → <code>03_calibration</code> → <code>04_threshold</code> → <code>05_feature_importance</code>) — 90 s</h3>
 <table class="bilingual">
 <tr>
 <td class="en">"Same model as last time — Random Forest, time-based split, τ = 0.20. Test ROC AUC <strong>0.871</strong>, PR AP <strong>0.750</strong>, Brier <strong>0.138</strong>, recall <strong>93.4 %</strong>. <strong>What's new is the 6 figures plus the model card</strong> — every number you see here is reproducible from <code>make evaluate</code>."</td>
 <td class="zh">"模型跟上次一样——RF、时间序列切分、τ = 0.20。测试 AUC <strong>0.871</strong>、PR AP <strong>0.750</strong>、Brier <strong>0.138</strong>、召回率 <strong>93.4%</strong>。<strong>新东西</strong>是 6 张图 + model card——上面任何一个数字都可以用 <code>make evaluate</code> 复现。"</td>
 </tr>
 </table>
 </div>
</div>
<div class="step">
 <div class="num">3</div>
 <div class="body">
 <h3>App (Tab <code>http://localhost:8000/app/</code>) — 60-90 s</h3>
 <table class="bilingual">
 <tr>
 <td class="en">"Step 3, the app — opened <strong>last</strong> as agreed. Two demo scenarios. First, <strong>Genting Highlands</strong> — a slope at 1865 m inside the training distribution. The model gives a moderate rain probability; the rule engine picks up orographic lift; the four mini-gauges decompose the risk by hazard type."</td>
 <td class="zh">"第三步 app——按约定<strong>最后才开</strong>。两个 demo 场景。第一个<strong>云顶高原</strong>——1865 m 的山坡，<strong>在训练分布之内</strong>。模型给中等降雨概率，规则引擎检测到地形抬升，四个 mini-gauge 把风险按灾害类型拆解。"</td>
 </tr>
 <tr>
 <td class="en">"Second, <strong>Mt Everest</strong> — completely out of distribution. The model alone would say 'safe'. The Veto cascade fires three independent overrides — hypoxia, frostbite, gale — and the composite is forced to Danger. There's a unit test for exactly this: <code>test_mt_everest_veto_hypoxia</code>."</td>
 <td class="zh">"第二个<strong>珠峰</strong>——<strong>完全分布外</strong>。光看模型会说"安全"，但 Veto 级联触发<strong>三个独立否决</strong>——缺氧、冻伤、大风——综合分被强制设为 Danger。<strong>专门有单元测试覆盖这个场景</strong>：<code>test_mt_everest_veto_hypoxia</code>。"</td>
 </tr>
 </table>
 </div>
</div>
<!-- ===== Section 4: Chapter 5 plan ===== -->
<h2 class="pb">4 · Next steps for Chapter 5
 <span class="zh">Chapter 5 下一步</span>
</h2>
<div class="callout warn">
 ≈ 90 seconds. <strong>This is the section the supervisor will react to most.</strong> Frame each item as a concrete deliverable + estimated time + dependency.<br>
 ≈ 90 秒。<strong>老师反应最强烈的就是这一节</strong>。每一项都以"<strong>交付物 + 估时 + 依赖</strong>"形式呈现。
</div>
<h3>4.1 · Proposed Chapter 5 work plan / Chapter 5 工作计划</h3>
<table class="plan">
 <tr><th>#</th><th>Deliverable / 交付物</th><th>EN one-liner</th><th>中文一句话</th><th class="estimate">Estimate</th></tr>
 <tr>
 <td><span class="pill plan">5.1</span></td>
 <td><strong>Comparative ablation</strong><br><span style="color:var(--muted);font-size:9pt">对比实验</span></td>
 <td>"Train LogReg + XGBoost on the same features and report ROC / PR / F2 side-by-side with RF — answers 'why RF?' empirically."</td>
 <td>"在同一特征集上训 LogReg + XGBoost，对比 ROC / PR / F2，<strong>用数据回答"为什么选 RF"</strong>。"</td>
 <td class="estimate">1 week</td>
 </tr>
 <tr>
 <td><span class="pill plan">5.2</span></td>
 <td><strong>Hindcast validation</strong><br><span style="color:var(--muted);font-size:9pt">历史事件回放</span></td>
 <td>"Replay 2020-2024 NaDMA-documented Malaysian flood / landslide events and check whether the system would have raised Warning / Danger at the right time. Reports hit-rate, lead-time, false-alarm rate."</td>
 <td>"把 2020-2024 NaDMA 公开的马来西亚洪水/滑坡事件<strong>逐一回放</strong>，看系统能否在事发前给出 Warning/Danger。报告命中率、提前量、误报率。"</td>
 <td class="estimate">2 weeks</td>
 </tr>
 <tr>
 <td><span class="pill plan">5.3</span></td>
 <td><strong>Threshold sensitivity</strong><br><span style="color:var(--muted);font-size:9pt">阈值灵敏度</span></td>
 <td>"Sweep τ ∈ {0.10, 0.15, 0.20, 0.25, 0.30}, plot precision-recall trade-off, and justify the operating point with a cost-of-error analysis."</td>
 <td>"扫 τ ∈ {0.10, 0.15, 0.20, 0.25, 0.30}，画精度-召回权衡曲线，用<strong>误差代价分析</strong>为最终选点辩护。"</td>
 <td class="estimate">3 days</td>
 </tr>
 <tr>
 <td><span class="pill plan">5.4</span></td>
 <td><strong>Component ablation</strong><br><span style="color:var(--muted);font-size:9pt">组件消融</span></td>
 <td>"Compare three system variants — RF only / Rule only / Hybrid — on the held-out test set and on the OOD Mt Everest case. Quantifies the rule-engine contribution."</td>
 <td>"对比三个系统变体——<strong>纯 RF / 纯规则 / 混合</strong>——在测试集和 OOD 珠峰场景上的表现。<strong>量化规则引擎的贡献</strong>。"</td>
 <td class="estimate">4 days</td>
 </tr>
 <tr>
 <td><span class="pill risk">5.5</span></td>
 <td><strong>Small user study</strong> <em>(optional)</em><br><span style="color:var(--muted);font-size:9pt">用户研究（可选）</span></td>
 <td>"Recruit 5-8 mountain hikers, run a 4-week panel, log system advice vs. their field judgment. Reports inter-rater agreement (Cohen's κ)."</td>
 <td>"招募 5-8 名登山者，4 周面板研究，记录系统建议 vs 他们现场判断，报告 Cohen's κ 一致性。"</td>
 <td class="estimate">4 weeks</td>
 </tr>
 <tr>
 <td><span class="pill done">5.6</span></td>
 <td><strong>Thesis Chapter 5 draft</strong><br><span style="color:var(--muted);font-size:9pt">章节初稿</span></td>
 <td>"Pull §5.1-5.5 into a single 12-15 page evaluation chapter with all figures, tables, and discussion."</td>
 <td>"把 §5.1-5.5 整合成 12-15 页的评估章节，含全部图表和讨论。"</td>
 <td class="estimate">1 week</td>
 </tr>
</table>
<h3>4.2 · Decision tree to ask the supervisor / 请示决策树</h3>
<div class="decision">
 <div class="decision-title">Q1 · Priorities</div>
 <table class="bilingual">
 <tr>
 <td class="en">"Sir, of the five evaluation tracks above, <strong>which two should I prioritise for the next four weeks</strong> before we converge on the Chapter 5 outline?"</td>
 <td class="zh">"老师，上面 5 条评估方向，<strong>未来四周</strong>您建议我重点做哪两条，然后再收敛到 Chapter 5 大纲？"</td>
 </tr>
 </table>
</div>
<div class="decision">
 <div class="decision-title">Q2 · User study yes/no</div>
 <table class="bilingual">
 <tr>
 <td class="en">"Do you want me to include the user study (5.5)? It is the longest item and depends on participant recruitment — I want your call before committing."</td>
 <td class="zh">"<strong>用户研究 (5.5) 您要不要做</strong>？这一条最长、依赖招募——想请您拍板再投入。"</td>
 </tr>
 </table>
</div>
<div class="decision">
 <div class="decision-title">Q3 · Framing of the comparative study</div>
 <table class="bilingual">
 <tr>
 <td class="en">"For the comparative ablation, do you want it framed as <strong>'why RF wins'</strong> (defending current choice) or <strong>'what if XGBoost wins'</strong> (open exploration)? The framing affects how I report inconclusive results."</td>
 <td class="zh">"<strong>对比实验</strong>您希望框成"为什么 RF 胜出"（<strong>捍卫现有选择</strong>）还是"如果 XGBoost 更好怎么办"（<strong>开放探索</strong>）？两种 framing 对<strong>模棱两可结果</strong>的报告方式不同。"</td>
 </tr>
 </table>
</div>
<div class="decision">
 <div class="decision-title">Q4 · Mt Everest weight in the thesis</div>
 <table class="bilingual">
 <tr>
 <td class="en">"Should I treat the Mt Everest OOD test as a <strong>thesis-level contribution</strong> (a stand-alone subsection on safety) or just an <strong>appendix item</strong>?"</td>
 <td class="zh">"<strong>珠峰 OOD 测试</strong>算论文级别的贡献（单独一节讲安全性），还是放附录就够？"</td>
 </tr>
 </table>
</div>
<!-- ===== Section 5: closing ===== -->
<h2 class="pb">5 · Asks + closing (60 seconds)
 <span class="zh">请示 + 收尾 60 秒</span>
</h2>
<table class="bilingual">
 <tr>
 <td class="en">"Sir, to summarise: since the last meeting I've shipped v1.0.0 — production-grade hardening, 70 tests at 97 % coverage, six evaluation figures, a published model card, full Docker reproducibility. The pipeline order is unchanged from what you asked: <strong>dataset, model, app</strong>. For Chapter 5 I have <strong>five evaluation tracks scoped</strong>; I'd like your guidance on which two to prioritise for the next four weeks."</td>
 <td class="zh">"老师，总结：自上次会议以来交付了 <strong>v1.0.0</strong>——工程化强化、70 个测试 97% 覆盖率、6 张评估图、model card、Docker 全复现。流水线顺序按您要求<strong>没动</strong>：dataset、model、app。Chapter 5 我列了 <strong>5 条评估方向</strong>，<strong>接下来四周您建议我先做哪两条</strong>？"</td>
 </tr>
 <tr>
 <td class="en">"I'll send you a 3-bullet email summary by tomorrow morning so we have <strong>written agreement</strong> on the priorities. Thank you for your time."</td>
 <td class="zh">"明早之前给您发 3 条要点的邮件总结，<strong>留个书面确认</strong>。谢谢老师。"</td>
 </tr>
</table>
<!-- ===== Section 6: defensive Q&A ===== -->
<h2>6 · Q&amp;A defensive lines (this update only)
 <span class="zh">本次进度汇报的兜底话术</span>
</h2>
<div class="callout tip">
 Anticipated follow-up questions <strong>specific to this progress update</strong>. The classic Q1-Q7 from the 5/11 brief are still live — just don't repeat them here.<br>
 <strong>针对本次进度汇报</strong>可能出现的追问。5/11 那份的经典 Q1-Q7 仍然有效，不重复罗列。
</div>
<h3>Q-N1 — "Why are you spending time on tests and Docker instead of the thesis?"</h3>
<h3 style="margin-top:-2mm">Q-N1 ——为什么你在写测试和 Docker 上花时间，不写论文？</h3>
<table class="bilingual">
 <tr>
 <td class="en">"Sir, the v1.0.0 hardening was a <strong>one-time investment</strong> to make every Chapter 5 number reproducible by the examiner with a single command. Without it, every evaluation result would be a black box — the examiner could not verify the AUC of 0.871 herself. With <code>make evaluate</code> reproducing all six figures byte-for-byte, the thesis claims become <strong>falsifiable</strong>. From this point on, all my time goes to evaluation and writing."</td>
 <td class="zh">"老师，v1.0.0 的强化是<strong>一次性投资</strong>——为了让评审老师<strong>用一行命令就能复现 Chapter 5 的每一个数字</strong>。没有它，AUC = 0.871 就是黑盒，<strong>评审无法独立验证</strong>。现在 <code>make evaluate</code> 能把 6 张图按字节复现，论文的每个 claim 都<strong>可证伪</strong>。从今天起所有时间都给评估和写作。"</td>
 </tr>
</table>
<h3>Q-N2 — "Why hasn't the model improved since last time?"</h3>
<h3 style="margin-top:-2mm">Q-N2 ——模型为什么自上次以后没提升？</h3>
<table class="bilingual">
 <tr>
 <td class="en">"Two reasons. First, the supervisor's instruction was to <em>consolidate</em> dataset and model before adding more capacity — which is what I did. Second, the bottleneck right now is <strong>not the model</strong> but the <strong>rule engine's coverage of OOD scenarios</strong>, which is a Chapter 5 contribution rather than a hyperparameter tweak. I'd rather report a defensible 0.871 with a calibrated rule engine than chase 0.88 with an unprincipled stack."</td>
 <td class="zh">"两个理由：(1) 您上次的指示是<strong>先把 dataset 和 model 巩固好</strong>再加复杂度——我严格照做了。(2) <strong>当前瓶颈不是模型本身</strong>，而是<strong>规则引擎对 OOD 场景的覆盖</strong>——这是 Chapter 5 的研究贡献，不是调超参。我宁愿报一个<strong>可辩护的 0.871</strong> 加一个校准好的规则引擎，<strong>也不要不讲原理地堆栈到 0.88</strong>。"</td>
 </tr>
</table>
<h3>Q-N3 — "Show me one concrete weakness you have not yet fixed."</h3>
<h3 style="margin-top:-2mm">Q-N3 ——给我说一个你目前还没修的具体弱点。</h3>
<table class="bilingual">
 <tr>
 <td class="en">"Honestly, Sir, the biggest one is <code>cape_jkg</code> — the ERA5 archive returns predominantly zero CAPE for these Malaysian coordinates, which is a <strong>known coverage gap</strong>. The Random Forest learns nothing from it (0 % importance). The rule engine still uses live Open-Meteo CAPE at inference time, so the production output is fine, but the <em>training</em> signal for thunderstorm risk is weaker than I'd like. I plan to address this in §5.4 ablation by quantifying how much it matters."</td>
 <td class="zh">"老实说，老师，最大的弱点是 <code>cape_jkg</code>——ERA5 在这些马来西亚坐标上的 CAPE 几乎全为零（<strong>已知覆盖缺口</strong>），<strong>RF 完全没学到东西</strong>（特征重要性 0%）。规则引擎在推理时用的是 Open-Meteo 实时 CAPE，所以生产输出没问题，但<strong>雷暴风险的训练信号</strong>比我希望的弱。计划在 §5.4 消融实验里<strong>量化它的影响</strong>。"</td>
 </tr>
</table>
<h3>Q-N4 — "When can I see the first draft of Chapter 5?"</h3>
<h3 style="margin-top:-2mm">Q-N4 ——Chapter 5 初稿什么时候能给我看？</h3>
<table class="bilingual">
 <tr>
 <td class="en">"If you sign off on tracks <strong>5.1 + 5.2 + 5.4</strong> today, the data collection finishes in 3 weeks, writing takes 1 week, so you'd have a draft in <strong>4 weeks from today</strong>. If you also want 5.5 (user study), add 4 weeks. <strong>I'll lock the date the moment you confirm the scope.</strong>"</td>
 <td class="zh">"如果今天您拍板 <strong>5.1 + 5.2 + 5.4</strong> 三条，<strong>3 周收数据 + 1 周写作 = 4 周后给您初稿</strong>。如果再加 <strong>5.5（用户研究）</strong>，再加 4 周。<strong>您一确认范围，我立刻锁定交稿日</strong>。"</td>
 </tr>
</table>
<!-- ===== Section 7: pre-flight checklist ===== -->
<h2 class="pb">7 · Pre-flight checklist (T-60 sec)
 <span class="zh">起飞前 60 秒自检</span>
</h2>
<div class="check">
<pre><code>☐ Laptop ≥ 80 % battery, charger in bag
☐ Terminal A: `make run` is running, do not close
☐ Terminal B: `curl /api/health` returned ml_loaded: true within last 5 min
☐ 10 browser tabs open in cheat-sheet §0 order — app tab is LAST
☐ This file open on a separate screen / phone, NOT to be read aloud
☐ docs/MEETING_CHEAT_SHEET.md open as a fall-back
☐ models/MODEL_CARD.md open in case any number is challenged
☐ figures/evaluation_summary.json downloadable on demand
☐ Phone on silent
☐ One deep breath. You shipped v1.0.0. You're prepared.</code></pre>
</div>
<div class="check">
<pre><code>☐ 笔记本电池 ≥ 80%，充电器已带
☐ 终端 A：`make run` 跑着，不要关
☐ 终端 B：5 分钟内 `curl /api/health` 返回 ml_loaded: true
☐ 10 个浏览器标签页按 cheat-sheet §0 顺序开好——app 标签放最后
☐ 本文档开在副屏 / 手机，不要照念
☐ docs/MEETING_CHEAT_SHEET.md 开着兜底
☐ models/MODEL_CARD.md 开着，老师质疑任何数字立刻打开
☐ figures/evaluation_summary.json 随时可发
☐ 手机静音
☐ 深呼吸。v1.0.0 已经交付。你准备好了。</code></pre>
</div>
<!-- ===== Section 8: cross references ===== -->
<h2>8 · Cross-references
 <span class="zh">相关文档索引</span>
</h2>
<table>
 <tr><th>Topic / 主题</th><th>File / 文件</th></tr>
 <tr><td>Original 5/11 reply to 4/15 feedback</td><td><a href="supervisor_meeting_brief.md"><code>supervisor_meeting_brief.md</code></a></td></tr>
 <tr><td>One-page cheat sheet (tab order, demo script)</td><td><a href="MEETING_CHEAT_SHEET.html"><code>MEETING_CHEAT_SHEET.html</code></a></td></tr>
 <tr><td>Pipeline order ASCII chart</td><td><a href="pipeline_order.md"><code>pipeline_order.md</code></a></td></tr>
 <tr><td>Dataset spec + Y derivation</td><td><a href="dataset.md"><code>dataset.md</code></a></td></tr>
 <tr><td>Architecture deep-dive</td><td><a href="architecture.md"><code>architecture.md</code></a></td></tr>
 <tr><td>Threshold citations</td><td><a href="thresholds.md"><code>thresholds.md</code></a></td></tr>
 <tr><td>Model card</td><td><a href="../models/MODEL_CARD.md"><code>../models/MODEL_CARD.md</code></a></td></tr>
 <tr><td>Evaluation summary JSON</td><td><a href="../figures/evaluation_summary.json"><code>../figures/evaluation_summary.json</code></a></td></tr>
 <tr><td>What changed in v1.0.0</td><td><a href="../CHANGELOG.md"><code>../CHANGELOG.md</code></a></td></tr>
</table>
<footer>
 Generated 2026-05-13 for the MicroClimate-X progress-update meeting at UKM.<br>
 此页为 2026-05-13 UKM 毕业设计 MicroClimate-X 进度汇报准备文档。<br>
 <span style="color: var(--brand);">L.ZH @ UKM · KyoukoLi/microclimate-x</span>
</footer>
</main>
</body>
</html>