flykhan
2536c937e3
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
98 lines
7.6 KiB
XML
98 lines
7.6 KiB
XML
<svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 800 280" width="800" height="280">
|
|
<defs>
|
|
<marker id="arr" markerWidth="8" markerHeight="6" refX="8" refY="3" orient="auto">
|
|
<path d="M0,0 L8,3 L0,6" fill="#333"/>
|
|
</marker>
|
|
<marker id="arr-purple" markerWidth="8" markerHeight="6" refX="8" refY="3" orient="auto">
|
|
<path d="M0,0 L8,3 L0,6" fill="#9b59b6"/>
|
|
</marker>
|
|
<marker id="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="400" y="22" text-anchor="middle" font-family="Arial, sans-serif" font-size="14" font-weight="bold" fill="#333">DALL-E: Autoregressive Image Generation</text>
|
|
|
|
<!-- Text tokens (red squares) -->
|
|
<text x="45" y="58" text-anchor="middle" font-family="Arial, sans-serif" font-size="10" font-weight="bold" fill="#e74c3c">Text Tokens</text>
|
|
<rect x="15" y="66" width="28" height="28" rx="4" fill="#e74c3c" fill-opacity="0.15" stroke="#e74c3c" stroke-width="1.2"/>
|
|
<text x="29" y="84" text-anchor="middle" font-family="Arial, sans-serif" font-size="9" fill="#e74c3c">a</text>
|
|
<rect x="47" y="66" width="28" height="28" rx="4" fill="#e74c3c" fill-opacity="0.15" stroke="#e74c3c" stroke-width="1.2"/>
|
|
<text x="61" y="84" text-anchor="middle" font-family="Arial, sans-serif" font-size="9" fill="#e74c3c">cat</text>
|
|
<rect x="79" y="66" width="28" height="28" rx="4" fill="#e74c3c" fill-opacity="0.15" stroke="#e74c3c" stroke-width="1.2"/>
|
|
<text x="93" y="84" text-anchor="middle" font-family="Arial, sans-serif" font-size="9" fill="#e74c3c">on</text>
|
|
<rect x="111" y="66" width="28" height="28" rx="4" fill="#e74c3c" fill-opacity="0.15" stroke="#e74c3c" stroke-width="1.2"/>
|
|
<text x="125" y="84" text-anchor="middle" font-family="Arial, sans-serif" font-size="9" fill="#e74c3c">mat</text>
|
|
|
|
<!-- Image tokens (blue squares) -->
|
|
<text x="205" y="58" text-anchor="middle" font-family="Arial, sans-serif" font-size="10" font-weight="bold" fill="#3498db">Image Tokens</text>
|
|
<rect x="155" y="66" width="28" height="28" rx="4" fill="#3498db" fill-opacity="0.15" stroke="#3498db" stroke-width="1.2"/>
|
|
<text x="169" y="84" text-anchor="middle" font-family="Arial, sans-serif" font-size="8" fill="#3498db">42</text>
|
|
<rect x="187" y="66" width="28" height="28" rx="4" fill="#3498db" fill-opacity="0.15" stroke="#3498db" stroke-width="1.2"/>
|
|
<text x="201" y="84" text-anchor="middle" font-family="Arial, sans-serif" font-size="8" fill="#3498db">17</text>
|
|
<rect x="219" y="66" width="28" height="28" rx="4" fill="#3498db" fill-opacity="0.15" stroke="#3498db" stroke-width="1.2"/>
|
|
<text x="233" y="84" text-anchor="middle" font-family="Arial, sans-serif" font-size="8" fill="#3498db">91</text>
|
|
<rect x="251" y="66" width="28" height="28" rx="4" fill="#3498db" fill-opacity="0.15" stroke="#3498db" stroke-width="1.2"/>
|
|
<text x="265" y="84" text-anchor="middle" font-family="Arial, sans-serif" font-size="8" fill="#3498db">...</text>
|
|
<rect x="283" y="66" width="28" height="28" rx="4" fill="#3498db" fill-opacity="0.15" stroke="#3498db" stroke-width="1.2"/>
|
|
<text x="297" y="84" text-anchor="middle" font-family="Arial, sans-serif" font-size="8" fill="#3498db">?</text>
|
|
|
|
<!-- Concatenation bracket -->
|
|
<text x="165" y="110" text-anchor="middle" font-family="Arial, sans-serif" font-size="9" fill="#666">concatenated sequence</text>
|
|
<line x1="15" y1="100" x2="310" y2="100" stroke="#ccc" stroke-width="1" stroke-dasharray="3,2"/>
|
|
|
|
<!-- Arrow down to transformer -->
|
|
<line x1="165" y1="118" x2="165" y2="138" stroke="#333" stroke-width="1.2" marker-end="url(#arr)"/>
|
|
|
|
<!-- Transformer Decoder -->
|
|
<rect x="70" y="140" width="190" height="55" rx="8" fill="#9b59b6" fill-opacity="0.12" stroke="#9b59b6" stroke-width="1.5"/>
|
|
<text x="165" y="163" text-anchor="middle" font-family="Arial, sans-serif" font-size="11" font-weight="bold" fill="#9b59b6">Transformer Decoder</text>
|
|
<text x="165" y="180" text-anchor="middle" font-family="Arial, sans-serif" font-size="9" fill="#666">12B parameters, causal attention</text>
|
|
|
|
<!-- Autoregressive arrow (curved, looping) -->
|
|
<path d="M260,167 C290,167 290,130 310,100 C320,85 310,75 297,80" fill="none" stroke="#9b59b6" stroke-width="1.2" stroke-dasharray="4,2" marker-end="url(#arr-purple)"/>
|
|
<text x="318" y="130" text-anchor="start" font-family="Arial, sans-serif" font-size="8" fill="#9b59b6">predict next</text>
|
|
<text x="318" y="140" text-anchor="start" font-family="Arial, sans-serif" font-size="8" fill="#9b59b6">image token</text>
|
|
|
|
<!-- Arrow from transformer to predicted tokens -->
|
|
<line x1="165" y1="195" x2="165" y2="215" stroke="#333" stroke-width="1.2" marker-end="url(#arr)"/>
|
|
|
|
<!-- Predicted image tokens row -->
|
|
<text x="165" y="233" text-anchor="middle" font-family="Arial, sans-serif" font-size="10" fill="#666">Predicted image token sequence (1024 tokens)</text>
|
|
|
|
<!-- Predicted tokens -->
|
|
<rect x="70" y="240" width="22" height="22" rx="3" fill="#3498db" fill-opacity="0.25" stroke="#3498db" stroke-width="1"/>
|
|
<rect x="96" y="240" width="22" height="22" rx="3" fill="#3498db" fill-opacity="0.25" stroke="#3498db" stroke-width="1"/>
|
|
<rect x="122" y="240" width="22" height="22" rx="3" fill="#3498db" fill-opacity="0.25" stroke="#3498db" stroke-width="1"/>
|
|
<rect x="148" y="240" width="22" height="22" rx="3" fill="#3498db" fill-opacity="0.25" stroke="#3498db" stroke-width="1"/>
|
|
<rect x="174" y="240" width="22" height="22" rx="3" fill="#3498db" fill-opacity="0.25" stroke="#3498db" stroke-width="1"/>
|
|
<rect x="200" y="240" width="22" height="22" rx="3" fill="#3498db" fill-opacity="0.25" stroke="#3498db" stroke-width="1"/>
|
|
<rect x="226" y="240" width="22" height="22" rx="3" fill="#3498db" fill-opacity="0.25" stroke="#3498db" stroke-width="1"/>
|
|
<text x="260" y="255" text-anchor="start" font-family="Arial, sans-serif" font-size="10" fill="#3498db">...</text>
|
|
|
|
<!-- Arrow to dVAE Decoder -->
|
|
<line x1="280" y1="251" x2="410" y2="251" stroke="#333" stroke-width="1.2" marker-end="url(#arr)"/>
|
|
|
|
<!-- dVAE Decoder -->
|
|
<rect x="420" y="125" width="140" height="50" rx="8" fill="#27ae60" fill-opacity="0.12" stroke="#27ae60" stroke-width="1.5"/>
|
|
<text x="490" y="148" text-anchor="middle" font-family="Arial, sans-serif" font-size="11" font-weight="bold" fill="#27ae60">dVAE Decoder</text>
|
|
<text x="490" y="163" text-anchor="middle" font-family="Arial, sans-serif" font-size="9" fill="#666">8192-vocab codebook</text>
|
|
|
|
<!-- Arrow connecting predicted tokens to dVAE -->
|
|
<line x1="310" y1="251" x2="420" y2="160" stroke="#333" stroke-width="1.2" marker-end="url(#arr)"/>
|
|
|
|
<!-- Arrow from dVAE to image -->
|
|
<line x1="560" y1="150" x2="610" y2="150" stroke="#333" stroke-width="1.2" marker-end="url(#arr)"/>
|
|
|
|
<!-- Output image placeholder -->
|
|
<rect x="620" y="100" width="100" height="100" rx="8" fill="#3498db" fill-opacity="0.08" stroke="#3498db" stroke-width="1.5"/>
|
|
<!-- Simple cat silhouette hint -->
|
|
<text x="670" y="145" text-anchor="middle" font-family="Arial, sans-serif" font-size="10" fill="#3498db">Generated</text>
|
|
<text x="670" y="158" text-anchor="middle" font-family="Arial, sans-serif" font-size="10" fill="#3498db">Image</text>
|
|
<text x="670" y="175" text-anchor="middle" font-family="Arial, sans-serif" font-size="9" fill="#999">256 x 256</text>
|
|
|
|
<!-- Stage labels at bottom -->
|
|
<text x="165" y="275" text-anchor="middle" font-family="Arial, sans-serif" font-size="9" fill="#999">Stage 1: Autoregressive token prediction</text>
|
|
<text x="560" y="275" text-anchor="middle" font-family="Arial, sans-serif" font-size="9" fill="#999">Stage 2: Token-to-pixel decoding</text>
|
|
</svg> |