Files
maths-cs-ai-compendium-zh/images/instructpix2pix_pipeline.svg
T
flykhan 2536c937e3 feat: 完整中文翻译 maths-cs-ai-compendium(数学·计算机科学·AI 知识大全)
翻译自英文原版 maths-cs-ai-compendium,共 20 章全部完成。

第01章 向量 | 第02章 矩阵 | 第03章 微积分
第04章 统计学 | 第05章 概率论 | 第06章 机器学习
第07章 计算语言学 | 第08章 计算机视觉 | 第09章 音频与语音
第10章 多模态学习 | 第11章 自主系统 | 第12章 图神经网络
第13章 计算与操作系统 | 第14章 数据结构与算法
第15章 生产级软件工程 | 第16章 SIMD与GPU编程
第17章 AI推理 | 第18章 ML系统设计
第19章 应用人工智能 | 第20章 前沿人工智能

翻译说明:
- 所有数学公式 $...$ / $$...$$、代码块、图片引用完整保留
- mkdocs.yml 配置中文导航 + language: zh
- README.md 已翻译为中文(兼 docs/index.md)
- docs/ 目录包含指向各章文件的 symlink
- 约 29,000 行中文内容,排除 .cache/ 构建缓存
2026-05-03 10:23:20 +08:00

94 lines
6.9 KiB
XML

<svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 750 280" width="750" height="280">
<defs>
<marker id="ip-arr" markerWidth="8" markerHeight="6" refX="8" refY="3" orient="auto">
<path d="M0,0 L8,3 L0,6" fill="#333"/>
</marker>
<marker id="ip-arr-red" markerWidth="8" markerHeight="6" refX="8" refY="3" orient="auto">
<path d="M0,0 L8,3 L0,6" fill="#e74c3c"/>
</marker>
<marker id="ip-arr-blue" markerWidth="8" markerHeight="6" refX="8" refY="3" orient="auto">
<path d="M0,0 L8,3 L0,6" fill="#3498db"/>
</marker>
</defs>
<!-- Title -->
<text x="375" y="22" text-anchor="middle" font-family="Arial, sans-serif" font-size="14" font-weight="bold" fill="#333">InstructPix2Pix: Instruction-Following Image Editing</text>
<!-- Input Image -->
<rect x="20" y="65" width="90" height="75" rx="8" fill="#3498db" fill-opacity="0.08" stroke="#3498db" stroke-width="1.5"/>
<!-- Simple scene: horizon line, sun circle -->
<line x1="30" y1="110" x2="100" y2="110" stroke="#3498db" stroke-width="0.8" opacity="0.4"/>
<circle cx="65" cy="90" r="10" fill="#3498db" fill-opacity="0.15" stroke="#3498db" stroke-width="0.8"/>
<rect x="45" y="110" width="15" height="20" rx="2" fill="#3498db" fill-opacity="0.12" stroke="#3498db" stroke-width="0.5"/>
<rect x="70" y="105" width="12" height="25" rx="2" fill="#3498db" fill-opacity="0.12" stroke="#3498db" stroke-width="0.5"/>
<text x="65" y="155" text-anchor="middle" font-family="Arial, sans-serif" font-size="9" font-weight="bold" fill="#3498db">Input Image</text>
<text x="65" y="167" text-anchor="middle" font-family="Arial, sans-serif" font-size="8" fill="#999">(daytime scene)</text>
<!-- Text Instruction -->
<rect x="20" y="185" width="105" height="35" rx="8" fill="#e74c3c" fill-opacity="0.08" stroke="#e74c3c" stroke-width="1.2" stroke-dasharray="4,2"/>
<text x="72" y="202" text-anchor="middle" font-family="Arial, sans-serif" font-size="9" fill="#e74c3c">"Make it sunset"</text>
<text x="72" y="214" text-anchor="middle" font-family="Arial, sans-serif" font-size="8" fill="#666">Text Instruction</text>
<!-- Arrows from inputs to model -->
<line x1="110" y1="102" x2="195" y2="115" stroke="#3498db" stroke-width="1.2" marker-end="url(#ip-arr-blue)"/>
<line x1="125" y1="202" x2="195" y2="145" stroke="#e74c3c" stroke-width="1.2" marker-end="url(#ip-arr-red)"/>
<!-- Modified Diffusion Model (large purple box) -->
<rect x="200" y="70" width="250" height="110" rx="8" fill="#9b59b6" fill-opacity="0.12" stroke="#9b59b6" stroke-width="1.5"/>
<text x="325" y="95" text-anchor="middle" font-family="Arial, sans-serif" font-size="11" font-weight="bold" fill="#9b59b6">Modified Diffusion Model</text>
<text x="325" y="112" text-anchor="middle" font-family="Arial, sans-serif" font-size="9" fill="#666">Conditioned on both image + text</text>
<!-- Inside the model: dual conditioning -->
<rect x="215" y="120" width="90" height="22" rx="4" fill="#3498db" fill-opacity="0.1" stroke="#3498db" stroke-width="0.8"/>
<text x="260" y="134" text-anchor="middle" font-family="Arial, sans-serif" font-size="8" fill="#3498db">Image concat</text>
<rect x="320" y="120" width="115" height="22" rx="4" fill="#e74c3c" fill-opacity="0.1" stroke="#e74c3c" stroke-width="0.8"/>
<text x="377" y="134" text-anchor="middle" font-family="Arial, sans-serif" font-size="8" fill="#e74c3c">Text cross-attention</text>
<!-- CFG formula -->
<text x="325" y="160" text-anchor="middle" font-family="Arial, sans-serif" font-size="8" fill="#666">CFG: e = e_u + s_i(e_i - e_u) + s_t(e_it - e_i)</text>
<!-- Guidance scales (sliders) -->
<text x="260" y="197" text-anchor="middle" font-family="Arial, sans-serif" font-size="9" font-weight="bold" fill="#333">Guidance Scales</text>
<!-- Image guidance slider -->
<text x="218" y="218" text-anchor="start" font-family="Arial, sans-serif" font-size="8" fill="#3498db">Image s_i</text>
<rect x="270" y="210" width="80" height="8" rx="4" fill="#ccc" fill-opacity="0.4" stroke="#ccc" stroke-width="0.5"/>
<rect x="270" y="210" width="50" height="8" rx="4" fill="#3498db" fill-opacity="0.4"/>
<circle cx="320" cy="214" r="5" fill="#3498db" stroke="white" stroke-width="1"/>
<text x="355" y="218" text-anchor="start" font-family="Arial, sans-serif" font-size="7" fill="#999">1.5</text>
<!-- Text guidance slider -->
<text x="218" y="240" text-anchor="start" font-family="Arial, sans-serif" font-size="8" fill="#e74c3c">Text s_t</text>
<rect x="270" y="232" width="80" height="8" rx="4" fill="#ccc" fill-opacity="0.4" stroke="#ccc" stroke-width="0.5"/>
<rect x="270" y="232" width="60" height="8" rx="4" fill="#e74c3c" fill-opacity="0.4"/>
<circle cx="330" cy="236" r="5" fill="#e74c3c" stroke="white" stroke-width="1"/>
<text x="355" y="240" text-anchor="start" font-family="Arial, sans-serif" font-size="7" fill="#999">7.5</text>
<!-- Arrow from model to output -->
<line x1="450" y1="125" x2="510" y2="125" stroke="#333" stroke-width="1.2" marker-end="url(#ip-arr)"/>
<!-- Output Image -->
<rect x="520" y="65" width="90" height="75" rx="8" fill="#f39c12" fill-opacity="0.08" stroke="#f39c12" stroke-width="1.5"/>
<!-- Same scene but sunset colors -->
<line x1="530" y1="110" x2="600" y2="110" stroke="#f39c12" stroke-width="0.8" opacity="0.4"/>
<circle cx="565" cy="90" r="10" fill="#f39c12" fill-opacity="0.2" stroke="#f39c12" stroke-width="0.8"/>
<!-- Sunset gradient hint -->
<rect x="525" y="70" width="80" height="40" rx="4" fill="#e74c3c" fill-opacity="0.05"/>
<rect x="545" y="110" width="15" height="20" rx="2" fill="#f39c12" fill-opacity="0.12" stroke="#f39c12" stroke-width="0.5"/>
<rect x="570" y="105" width="12" height="25" rx="2" fill="#f39c12" fill-opacity="0.12" stroke="#f39c12" stroke-width="0.5"/>
<text x="565" y="155" text-anchor="middle" font-family="Arial, sans-serif" font-size="9" font-weight="bold" fill="#f39c12">Edited Image</text>
<text x="565" y="167" text-anchor="middle" font-family="Arial, sans-serif" font-size="8" fill="#999">(sunset scene)</text>
<!-- Preservation note -->
<rect x="520" y="185" width="200" height="45" rx="6" fill="#27ae60" fill-opacity="0.06" stroke="#27ae60" stroke-width="1" stroke-dasharray="4,2"/>
<text x="620" y="202" text-anchor="middle" font-family="Arial, sans-serif" font-size="9" font-weight="bold" fill="#27ae60">Structure Preserved</text>
<text x="620" y="216" text-anchor="middle" font-family="Arial, sans-serif" font-size="8" fill="#666">Same layout, buildings, composition</text>
<text x="620" y="228" text-anchor="middle" font-family="Arial, sans-serif" font-size="8" fill="#666">Only lighting/color changed per instruction</text>
<!-- Connecting arrow from output to preservation note -->
<line x1="565" y1="167" x2="580" y2="185" stroke="#27ae60" stroke-width="0.8" stroke-dasharray="3,2"/>
<!-- Bottom annotation -->
<text x="375" y="270" text-anchor="middle" font-family="Arial, sans-serif" font-size="9" fill="#999">No per-example fine-tuning required -- single forward pass at inference</text>
</svg>