├── .gitignore
├── Gemfile
├── README.md
├── _announcements
    ├── week-01.md
    ├── week-02.md
    ├── week-04.md
    ├── week-05.md
    ├── week-06.md
    ├── week-11.md
    ├── week-16.md
    ├── week-18.md
    ├── week-21.md
    ├── week-22.md
    └── week-24.md
├── _config.yml
├── _includes
    ├── head_custom.html
    └── minutes.liquid
├── _layouts
    ├── announcement.html
    ├── minimal.html
    ├── module.html
    ├── schedule.html
    └── staffer.html
├── _modules
    ├── part_0.json
    ├── part_0.md
    ├── part_1.json
    ├── part_1.md
    ├── part_2.json
    ├── part_2.md
    ├── part_3.json
    ├── part_3.md
    ├── part_4.json
    └── part_4.md
├── _sass
    ├── color_schemes
    │   └── d2l.scss
    └── custom
    │   ├── announcement.scss
    │   ├── card.scss
    │   ├── custom.scss
    │   ├── module.scss
    │   ├── schedule.scss
    │   └── staffer.scss
├── _staffers
    └── mu.md
├── announcements.md
├── assets
    ├── images
    │   ├── logo-with-text.png
    │   └── mu.jpg
    └── pdfs
    │   ├── part-0_1.pdf
    │   ├── part-0_2.pdf
    │   ├── part-0_3.pdf
    │   ├── part-0_4.pdf
    │   ├── part-0_5.pdf
    │   ├── part-0_6.pdf
    │   ├── part-0_7.pdf
    │   └── part-0_8.pdf
├── contents.json
├── favicon.ico
├── generate.ipynb
├── index.md
└── upload.sh


/.gitignore:
--------------------------------------------------------------------------------
 1 | _site/
 2 | .sass-cache/
 3 | .jekyll-cache/
 4 | .jekyll-metadata
 5 | .bundle/
 6 | **/.DS_Store
 7 | Gemfile.lock
 8 | vendor
 9 | **/pdfs/
10 | **/notebooks/
11 | *.ipynb_checkpoints/


--------------------------------------------------------------------------------
/Gemfile:
--------------------------------------------------------------------------------
1 | source 'https://rubygems.org'
2 | gem "jekyll"
3 | gem "just-the-docs"
4 | gem "webrick"


--------------------------------------------------------------------------------
/README.md:
--------------------------------------------------------------------------------
 1 | # 《动手学深度学习》v2 课程
 2 | 
 3 | 课程主页：https://courses.d2l.ai/zh-v2/
 4 | 
 5 | 本地编译
 6 | 
 7 | ```bash
 8 | bundle install
 9 | bundle exec jekyll serve
10 | ```
11 | 
12 | 


--------------------------------------------------------------------------------
/_announcements/week-01.md:
--------------------------------------------------------------------------------
1 | ---
2 | title: 第一周
3 | week: 1
4 | ---
5 | 
6 | 第一周视频上传到了B站。根据大家建议，将相关的单元做成了合集。每个视频下面有1到4个分集。可以通过[B站频道](https://space.bilibili.com/1567748478/channel/index)来按序访问，或者按照下面课程安排里的视频连接来访问。此外，因为直播时“按特定轴求和”没有讲得很清楚，直播后补录了一个视频。


--------------------------------------------------------------------------------
/_announcements/week-02.md:
--------------------------------------------------------------------------------
1 | ---
2 | title: 第二周
3 | week: 2
4 | ---
5 | 
6 | 第二周视频上传到了B站。其中线性回归一节是直播后重录，会跟直播内容有点点区别。请注意下周因为清明直播停一周。


--------------------------------------------------------------------------------
/_announcements/week-04.md:
--------------------------------------------------------------------------------
1 | ---
2 | title: 第四周
3 | week: 4
4 | ---
5 | 
6 | 第一部分完结。完结福利是第一次课程竞赛，内容为[预测2020加州房价](https://www.kaggle.com/c/california-house-prices/overview)。前5个人（团队）将获得作者签名版《动手学深度学习》第二版。介绍视频请见[B站](https://www.bilibili.com/video/BV1NK4y1P7Tu?p=2)。


--------------------------------------------------------------------------------
/_announcements/week-05.md:
--------------------------------------------------------------------------------
1 | ---
2 | title: 第五周
3 | week: 5
4 | ---
5 | 
6 | 因为五一长假关系，4月25日、5月1日、和5月2日停课。[课程竞赛](https://www.kaggle.com/c/california-house-prices/overview)截止日期也延期到5月9日直播前1小时。
7 | 


--------------------------------------------------------------------------------
/_announcements/week-06.md:
--------------------------------------------------------------------------------
1 | ---
2 | title: 第六周
3 | week: 6
4 | ---
5 | 
6 | 因为5月8日调休，考虑到单休周大家可能事会比较多，所以这周仍然停课。课程竞赛截止日期不变，仍然是5月9号中午12:00（也就是kaggle上显示时间）。周日会在B站介绍基线automl模型。
7 | 


--------------------------------------------------------------------------------
/_announcements/week-11.md:
--------------------------------------------------------------------------------
1 | ---
2 | title: 第十一周
3 | week: 11
4 | ---
5 | 
6 | 第二部分卷积神经网络完结。录像全部上传到了[B站](https://space.bilibili.com/1567748478/channel/detail?cid=175509)。本部分竞赛为树叶分类，[Kaggle地址](https://www.kaggle.com/c/classify-leaves)
7 | 


--------------------------------------------------------------------------------
/_announcements/week-16.md:
--------------------------------------------------------------------------------
1 | ---
2 | title: 第十六周 
3 | week: 16
4 | ---
5 | 
6 | 树叶分类竞赛结束，查看[获胜团队和技术总结](https://www.bilibili.com/video/BV1by4y1K7SE)
7 | 


--------------------------------------------------------------------------------
/_announcements/week-18.md:
--------------------------------------------------------------------------------
1 | ---
2 | title: 第十八周 
3 | week: 18
4 | ---
5 | 
6 | 第三部分计算机视觉结束，第三次竞赛为[牛仔行头检测](https://www.bilibili.com/video/BV1F64y1x7xP/)
7 | 


--------------------------------------------------------------------------------
/_announcements/week-21.md:
--------------------------------------------------------------------------------
1 | ---
2 | title: 第二十一周 
3 | week: 21
4 | ---
5 | 
6 | 因为出差原因，本周末（7月31日和8月1日）课程直播取消。
7 | 


--------------------------------------------------------------------------------
/_announcements/week-22.md:
--------------------------------------------------------------------------------
1 | ---
2 | title: 第二十二周 
3 | week: 22
4 | ---
5 | 
6 | 目标检测竞赛结束。[前十名同学](https://competitions.codalab.org/competitions/33573#results)（lcpdeb，herunyu，signcoda，zZamm，sophiezang，saltFish，snowtyan，nekokiku，dejahu，momo233）如果在8月15日前提交代码到Kaggle，将获得签名赠书。
7 | 


--------------------------------------------------------------------------------
/_announcements/week-24.md:
--------------------------------------------------------------------------------
1 | ---
2 | title: 第二十四周 
3 | week: 24
4 | ---
5 | 
6 | 课程结束！谢谢学习。所有视频均上传[B站](https://space.bilibili.com/1567748478/channel/detail?cid=175509)。欢迎观看接下来的课程：[斯坦福2021秋季：实用机器学习](https://c.d2l.ai/stanford-cs329p/)。
7 | 


--------------------------------------------------------------------------------
/_config.yml:
--------------------------------------------------------------------------------
 1 | title: "动手学深度学习课程"
 2 | theme: just-the-docs
 3 | remote_theme: pmarsceill/just-the-docs
 4 | 
 5 | logo: "/assets/images/logo-with-text.png"
 6 | 
 7 | aux_links:
 8 |   "GitHub":
 9 |     - "//github.com/mli/course-zh-v2"
10 | aux_links_new_tab: true
11 | 
12 | heading_anchors: true
13 | 
14 | ga_tracking: UA-96378503-2
15 | ga_tracking_anonymize_ip: true
16 | 
17 | color_scheme: d2l
18 | 
19 | 
20 | tagline: 动手学深度学习在线课程
21 | description: 免费深度学习在线课程
22 | author: 李沐
23 | baseurl: '/zh-v2' # the subpath of your site, e.g. /blog
24 | url: 'https://courses.d2l.ai'
25 | exclude: ["Gemfile", "Gemfile.lock", "LICENSE", "README.md"]
26 | 
27 | # Collections for website data
28 | collections:
29 |   staffers:
30 |   modules:
31 |   announcements:
32 | 
33 | # Default layouts for each collection type
34 | defaults:
35 |   - scope:
36 |       path: ''
37 |       type: staffers
38 |     values:
39 |       layout: staffer
40 |       subpath: '/assets/images/'
41 |   - scope:
42 |       path: ''
43 |       type: modules
44 |     values:
45 |       layout: module
46 |       subpath: '/assets/pdfs/'
47 |   - scope:
48 |       path: ''
49 |       type: announcements
50 |     values:
51 |       layout: announcement
52 | 
53 | compress_html:
54 |   clippings: all
55 |   comments: all
56 |   endings: all
57 |   startings: []
58 |   blanklines: false
59 |   profile: false
60 | 


--------------------------------------------------------------------------------
/_includes/head_custom.html:
--------------------------------------------------------------------------------
1 | <!-- https://cdnjs.com/libraries/font-awesome -->
2 | <!-- <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/5.15.2/css/all.min.css"> -->
3 | <link rel="stylesheet" href="https://at.alicdn.com/t/font_2371118_b27k2sys2hd.css">


--------------------------------------------------------------------------------
/_includes/minutes.liquid:
--------------------------------------------------------------------------------
 1 | {% capture _minutes_workspace %}
 2 |   {% comment %}
 3 |     Return the number of minutes between midnight and the given time string (e.g. '9:30 AM').
 4 | 
 5 |     Parameters:
 6 |       `time` (string): the time to convert.
 7 |   {% endcomment %}
 8 | 
 9 |   {% assign _time = include.time %}
10 |   {% assign _hhmm = _time | split: ' ' | first | split: ':' %}
11 |   {% assign _hours = _hhmm | first | to_integer %}
12 |   {% assign _minutes = _hhmm | last | to_integer %}
13 |   {% assign _ampm = _time | split: ' ' | last | upcase %}
14 | 
15 |   {% if _ampm == 'AM' and _hours == 12 %}
16 |     {% assign _hours = _hours | minus: 12 %}
17 |   {% elsif _ampm == 'PM' and _hours != 12 %}
18 |     {% assign _hours = _hours | plus: 12 %}
19 |   {% endif %}
20 | {% endcapture %}{% assign _minutes_workspace = '' %}{{ _hours | times: 60 | plus: _minutes }}
21 | 


--------------------------------------------------------------------------------
/_layouts/announcement.html:
--------------------------------------------------------------------------------
 1 | <div class="announcement">
 2 |   <h2>{{ page.title }}</h2>
 3 |   <span class="announcement-meta">
 4 |     {% assign minutes = content | strip_html | number_of_words | divided_by: 180.0 | round %}
 5 |   </span>
 6 |   <div>
 7 |     {{ content }}
 8 |   </div>
 9 | </div>
10 | 


--------------------------------------------------------------------------------
/_layouts/minimal.html:
--------------------------------------------------------------------------------
 1 | ---
 2 | layout: table_wrappers
 3 | ---
 4 | 
 5 | <!DOCTYPE html>
 6 | 
 7 | <html lang="{{ site.lang | default: 'en-US' }}">
 8 | {% include head.html %}
 9 | <body>
10 |   <svg xmlns="http://www.w3.org/2000/svg" style="display: none;">
11 |     <symbol id="svg-link" viewBox="0 0 24 24">
12 |       <title>Link</title>
13 |       <svg xmlns="http://www.w3.org/2000/svg" width="24" height="24" viewBox="0 0 24 24" fill="none" stroke="currentColor" stroke-width="2" stroke-linecap="round" stroke-linejoin="round" class="feather feather-link">
14 |         <path d="M10 13a5 5 0 0 0 7.54.54l3-3a5 5 0 0 0-7.07-7.07l-1.72 1.71"></path><path d="M14 11a5 5 0 0 0-7.54-.54l-3 3a5 5 0 0 0 7.07 7.07l1.71-1.71"></path>
15 |       </svg>
16 |     </symbol>
17 |   </svg>
18 | 
19 |   <div id="main-content-wrap" class="main-content-wrap">
20 |     {% unless page.url == "/" %}
21 |       {% if page.parent %}
22 |         <nav aria-label="Breadcrumb" class="breadcrumb-nav">
23 |           <ol class="breadcrumb-nav-list">
24 |             {% if page.grand_parent %}
25 |               <li class="breadcrumb-nav-list-item"><a href="{{ first_level_url }}">{{ page.grand_parent }}</a></li>
26 |               <li class="breadcrumb-nav-list-item"><a href="{{ second_level_url }}">{{ page.parent }}</a></li>
27 |             {% else %}
28 |               <li class="breadcrumb-nav-list-item"><a href="{{ first_level_url }}">{{ page.parent }}</a></li>
29 |             {% endif %}
30 |             <li class="breadcrumb-nav-list-item"><span>{{ page.title }}</span></li>
31 |           </ol>
32 |         </nav>
33 |       {% endif %}
34 |     {% endunless %}
35 |     <div id="main-content" class="main-content" role="main">
36 |       {% if site.heading_anchors != false %}
37 |         {% include vendor/anchor_headings.html html=content beforeHeading="true" anchorBody="<svg viewBox=\"0 0 16 16\" aria-hidden=\"true\"><use xlink:href=\"#svg-link\"></use></svg>" anchorClass="anchor-heading" %}
38 |       {% else %}
39 |         {{ content }}
40 |       {% endif %}
41 |     </div>
42 |   </div>
43 | </body>
44 | </html>
45 | 


--------------------------------------------------------------------------------
/_layouts/module.html:
--------------------------------------------------------------------------------
1 | <h2 class="fs-4" id="{{ page.title | slugify }}" >{{ page.title }}</h2>
2 | <div class="module">
3 |   {{ content }}
4 | </div>
5 | 


--------------------------------------------------------------------------------
/_layouts/schedule.html:
--------------------------------------------------------------------------------
 1 | {% assign start_time = page.timeline | first %}
 2 | {% capture offset %}{% include minutes.liquid time=start_time %}{% endcapture %}
 3 | <div class="schedule">
 4 |   <ul class="schedule-timeline" style="min-width: {{ page.schedule | size | times: 130 }}px">
 5 |     {% for time in page.timeline %}
 6 |     <li class="schedule-time">{{ time }} </li>
 7 |     {% endfor %}
 8 |   </ul>
 9 |   <ul class="schedule-group">
10 |     {% for day in page.schedule %}
11 |     <li class="schedule-day">
12 |       <h2 class="schedule-header">{{ day.name }}</h2>
13 |       {% if day.events %}
14 |       <ul class="schedule-events" style="height: {{ page.timeline | size | times: 40 }}px">
15 |       {% for event in day.events %}
16 |         {% capture start %}{% include minutes.liquid time=event.start %}{% endcapture %}
17 |         {% capture end %}{% include minutes.liquid time=event.end %}{% endcapture %}
18 |         {% assign top = start | minus: offset | times: 40 | divided_by: 30 %}
19 |         {% assign height = end | minus: start | times: 40 | divided_by: 30 %}
20 |         <li class="schedule-event {% if event.class %}{{ event.class }}{% else %}{{ event.name | slugify }}{% endif %}"
21 |             style="top: {{ top }}px; height: {{ height }}px;">
22 |           <div class="name">{{ event.name }}</div>
23 |           <div class="time">{{ event.start }}–{{ event.end }}</div>
24 |           {% if event.location %}
25 |           <div class="location">{{ event.location }}</div>
26 |           {% endif %}
27 |         </li>
28 |       {% endfor %}
29 |       </ul>
30 |       {% endif %}
31 |     </li>
32 |     {% endfor %}
33 |   </ul>
34 | </div>
35 | 


--------------------------------------------------------------------------------
/_layouts/staffer.html:
--------------------------------------------------------------------------------
 1 | <div class="staffer">
 2 |   {% if page.photo %}
 3 |   <img class="staffer-image" src="{{ site.baseurl }}{{ page.subpath }}{{ page.photo }}" alt="">
 4 |   {% endif %}
 5 |   <div class="staffer-info">
 6 |     <h3 class="staffer-name">
 7 |       {% if page.website %}
 8 |       <a href="{{ page.website }}">{{ page.name }}</a>
 9 |       {% else %}
10 |       {{ page.name }}
11 |       {% endif %}
12 |       {% if page.pronouns %}
13 |       <span class="staffer-pronouns">{{ page.pronouns }}</span>
14 |       {% endif %}
15 |     </h3>
16 |     {% if page.email %}
17 |     <p><a href="mailto:{{ page.email }}">{{ page.email }}</a></p>
18 |     {% endif %}
19 |     {% if page.section %}
20 |     <p class="staffer-meta">Quiz Section: {{ page.section | markdownify | strip_html }}</p>
21 |     {% endif %}
22 |     {% if page.office-hours %}
23 |     <p class="staffer-meta">Office Hours: {{ page.office-hours | markdownify | strip_html }}</p>
24 |     {% endif %}
25 |     {{ content }}
26 |   </div>
27 |   <div style="clear:both"></div>
28 | </div>


--------------------------------------------------------------------------------
/_modules/part_0.json:
--------------------------------------------------------------------------------
  1 | [
  2 |     {
  3 |         "title": "课程安排",
  4 |         "day_break": true,
  5 |         "slides": [
  6 |             "part-01.pdf",
  7 |             6
  8 |         ],
  9 |         "slides_video": "https://www.bilibili.com/video/BV1oX4y137bC",
 10 |         "notebook": false,
 11 |         "notebook_video": "",
 12 |         "qa_video": ""
 13 |     },
 14 |     {
 15 |         "title": "深度学习介绍",
 16 |         "day_break": false,
 17 |         "book": "chapter_introduction/index.html",
 18 |         "slides": [
 19 |             "part-01.pdf",
 20 |             14
 21 |         ],
 22 |         "slides_video": "https://www.bilibili.com/video/BV1J54y187f9",
 23 |         "notebook_video": ""
 24 |     },
 25 |     {
 26 |         "title": "安装",
 27 |         "day_break": false,
 28 |         "book": "chapter_installation/index.html",
 29 |         "slides": [
 30 |             "part-01.pdf",
 31 |             2
 32 |         ],
 33 |         "slides_video": "https://www.bilibili.com/video/BV18p4y1h7Dr",
 34 |         "qa_video": "https://www.bilibili.com/video/BV18p4y1h7Dr?p=2"
 35 |     },
 36 |     {
 37 |         "title": "数据操作",
 38 |         "day_break": false,
 39 |         "book": "chapter_preliminaries/ndarray.html",
 40 |         "slides": [
 41 |             "part-01.pdf",
 42 |             5
 43 |         ],
 44 |         "slides_video": "https://www.bilibili.com/video/BV1CV411Y7i4",
 45 |         "notebook_video": "https://www.bilibili.com/video/BV1CV411Y7i4?p=2",
 46 |         "qa_video": "https://www.bilibili.com/video/BV1CV411Y7i4?p=4"
 47 |     },
 48 |     {
 49 |         "title": "数据预处理",
 50 |         "day_break": false,
 51 |         "book": "chapter_preliminaries/pandas.html",
 52 |         "slides": [
 53 |             "part-01.pdf",
 54 |             0
 55 |         ],
 56 |         "slides_video": "https://www.bilibili.com/video/BV1CV411Y7i4?p=3",
 57 |         "notebook_video": "",
 58 |         "qa_video": ""
 59 |     },
 60 |     {
 61 |         "title": "线性代数",
 62 |         "day_break": true,
 63 |         "book": "chapter_preliminaries/linear-algebra.html",
 64 |         "slides": [
 65 |             "part-01.pdf",
 66 |             14
 67 |         ],
 68 |         "slides_video": "https://www.bilibili.com/video/BV1eK4y1U7Qy",
 69 |         "notebook_video": "https://www.bilibili.com/video/BV1eK4y1U7Qy?p=2",
 70 |         "qa_video": "https://www.bilibili.com/video/BV1eK4y1U7Qy?p=4"
 71 |     },
 72 |     {
 73 |         "title": "[补充] 按特定轴求和",
 74 |         "day_break": false,
 75 |         "slides": [
 76 |             "part-01.pdf",
 77 |             0
 78 |         ],
 79 |         "slides_video": "https://www.bilibili.com/video/BV1eK4y1U7Qy?p=3",
 80 |         "notebook_video": "",
 81 |         "qa_video": ""
 82 |     },
 83 |     {
 84 |         "title": "矩阵计算",
 85 |         "day_break": false,
 86 |         "book": "chapter_preliminaries/calculus.html",
 87 |         "slides": [
 88 |             "part-01.pdf",
 89 |             10
 90 |         ],
 91 |         "notebook": false,
 92 |         "slides_video": "https://www.bilibili.com/video/BV1eZ4y1w7PY",
 93 |         "notebook_video": "",
 94 |         "qa_video": "https://www.bilibili.com/video/BV1eZ4y1w7PY?p=2"
 95 |     },
 96 |     {
 97 |         "title": "自动求导",
 98 |         "day_break": false,
 99 |         "book": "chapter_preliminaries/autograd.html",
100 |         "slides": [
101 |             "part-01.pdf",
102 |             15
103 |         ],
104 |         "slides_video": "https://www.bilibili.com/video/BV1KA411N7Px",
105 |         "notebook_video": "https://www.bilibili.com/video/BV1KA411N7Px?p=2",
106 |         "qa_video": "https://www.bilibili.com/video/BV1KA411N7Px?p=3"
107 |     },
108 |     {
109 |         "title": "线性回归",
110 |         "day_break": true,
111 |         "book": "chapter_linear-networks/linear-regression.html",
112 |         "slides": [
113 |             "part-01.pdf",
114 |             11
115 |         ],
116 |         "slides_video": "https://www.bilibili.com/video/BV1PX4y1g7KC",
117 |         "notebook_video": "",
118 |         "qa_video": ""
119 |     },
120 |     {
121 |         "title": "基础优化方法",
122 |         "day_break": false,
123 |         "slides": [
124 |             "part-01.pdf",
125 |             6
126 |         ],
127 |         "slides_video": "https://www.bilibili.com/video/BV1PX4y1g7KC?p=2",
128 |         "notebook_video": "",
129 |         "qa_video": ""
130 |     },
131 |     {
132 |         "title": "线性回归的从零开始实现",
133 |         "day_break": false,
134 |         "book": "chapter_linear-networks/linear-regression-scratch.html",
135 |         "slides": [
136 |             "part-0.pdf",
137 |             0
138 |         ],
139 |         "slides_video": "https://www.bilibili.com/video/BV1PX4y1g7KC?p=3",
140 |         "notebook_video": "",
141 |         "qa_video": ""
142 |     },
143 |     {
144 |         "title": "线性回归的简洁实现",
145 |         "day_break": false,
146 |         "book": "chapter_linear-networks/linear-regression-concise.html",
147 |         "slides": [
148 |             "part-0.pdf",
149 |             0
150 |         ],
151 |         "slides_video": "https://www.bilibili.com/video/BV1PX4y1g7KC?p=4",
152 |         "notebook_video": "",
153 |         "qa_video": ""
154 |     },
155 |     {
156 |         "title": "Softmax 回归",
157 |         "day_break": true,
158 |         "book": "chapter_linear-networks/softmax-regression.html",
159 |         "slides": [
160 |             "part-01.pdf",
161 |             11
162 |         ],
163 |         "slides_video": "https://www.bilibili.com/video/BV1K64y1Q7wu",
164 |         "notebook_video": "",
165 |         "qa_video": ""
166 |     },
167 |     {
168 |         "title": "损失函数",
169 |         "day_break": false,
170 |         "book": "",
171 |         "slides": [
172 |             "part-01.pdf",
173 |             7
174 |         ],
175 |         "slides_video": "https://www.bilibili.com/video/BV1K64y1Q7wu?p=2",
176 |         "notebook_video": "",
177 |         "qa_video": ""
178 |     },
179 |     {
180 |         "title": "图像分类数据集",
181 |         "day_break": false,
182 |         "book": "chapter_linear-networks/image-classification-dataset.html",
183 |         "slides": [
184 |             "part-0.pdf",
185 |             0
186 |         ],
187 |         "slides_video": "https://www.bilibili.com/video/BV1K64y1Q7wu?p=3",
188 |         "notebook_video": "",
189 |         "qa_video": ""
190 |     },
191 |     {
192 |         "title": "Softmax 回归的从零开始实现",
193 |         "day_break": false,
194 |         "book": "chapter_linear-networks/softmax-regression-scratch.html",
195 |         "slides": [
196 |             "part-0.pdf",
197 |             0
198 |         ],
199 |         "slides_video": "https://www.bilibili.com/video/BV1K64y1Q7wu?p=4",
200 |         "notebook_video": "",
201 |         "qa_video": ""
202 |     },
203 |     {
204 |         "title": "Softmax 回归的简洁实现",
205 |         "day_break": false,
206 |         "book": "chapter_linear-networks/softmax-regression-concise.html",
207 |         "slides": [
208 |             "part-0.pdf",
209 |             0
210 |         ],
211 |         "slides_video": "https://www.bilibili.com/video/BV1K64y1Q7wu?p=5",
212 |         "notebook_video": "",
213 |         "qa_video": ""
214 |     },
215 |     {
216 |         "title": "感知机",
217 |         "day_break": true,
218 |         "slides": [
219 |             "part-01.pdf",
220 |             11
221 |         ],
222 |         "slides_video": "https://www.bilibili.com/video/BV1hh411U7gn",
223 |         "notebook_video": "",
224 |         "qa_video": ""
225 |     },
226 |     {
227 |         "title": "多层感知机",
228 |         "day_break": false,
229 |         "book": "chapter_multilayer-perceptrons/mlp.html",
230 |         "slides": [
231 |             "part-01.pdf",
232 |             13
233 |         ],
234 |         "slides_video": "https://www.bilibili.com/video/BV1hh411U7gn?p=2",
235 |         "notebook_video": "",
236 |         "qa_video": ""
237 |     },
238 |     {
239 |         "title": "多层感知机的从零开始实现",
240 |         "day_break": false,
241 |         "book": "chapter_multilayer-perceptrons/mlp-scratch.html",
242 |         "slides": [
243 |             "part-0.pdf",
244 |             0
245 |         ],
246 |         "slides_video": "https://www.bilibili.com/video/BV1hh411U7gn?p=3",
247 |         "notebook_video": "",
248 |         "qa_video": ""
249 |     },
250 |     {
251 |         "title": "多层感知机的简洁实现",
252 |         "day_break": false,
253 |         "book": "chapter_multilayer-perceptrons/mlp-concise.html",
254 |         "slides": [
255 |             "part-0.pdf",
256 |             0
257 |         ],
258 |         "slides_video": "https://www.bilibili.com/video/BV1hh411U7gn?p=3",
259 |         "notebook_video": "",
260 |         "qa_video": ""
261 |     },
262 |     {
263 |         "title": "模型选择",
264 |         "day_break": true,
265 |         "book": "",
266 |         "slides": [
267 |             "part-01.pdf",
268 |             7
269 |         ],
270 |         "slides_video": "https://www.bilibili.com/video/BV1kX4y1g7jp",
271 |         "notebook_video": "",
272 |         "qa_video": ""
273 |     },
274 |     {
275 |         "title": "欠拟合和过拟合",
276 |         "day_break": false,
277 |         "book": "chapter_multilayer-perceptrons/underfit-overfit.html",
278 |         "slides": [
279 |             "part-01.pdf",
280 |             10
281 |         ],
282 |         "slides_video": "https://www.bilibili.com/video/BV1kX4y1g7jp?p=2",
283 |         "notebook_video": "",
284 |         "qa_video": ""
285 |     },
286 |     {
287 |         "title": "权重衰减",
288 |         "day_break": true,
289 |         "book": "chapter_multilayer-perceptrons/weight-decay.html",
290 |         "slides": [
291 |             "part-01.pdf",
292 |             6
293 |         ],
294 |         "slides_video": "https://www.bilibili.com/video/BV1UK4y1o7dy",
295 |         "notebook_video": "",
296 |         "qa_video": ""
297 |     },
298 |     {
299 |         "title": "Dropout",
300 |         "day_break": false,
301 |         "book": "chapter_multilayer-perceptrons/dropout.html",
302 |         "slides": [
303 |             "part-01.pdf",
304 |             6
305 |         ],
306 |         "slides_video": "https://www.bilibili.com/video/BV1Y5411c7aY",
307 |         "notebook_video": "",
308 |         "qa_video": ""
309 |     },
310 |     {
311 |         "title": "数值稳定性",
312 |         "day_break": true,
313 |         "book": "chapter_multilayer-perceptrons/numerical-stability-and-init.html",
314 |         "slides": [
315 |             "part-01.pdf",
316 |             10
317 |         ],
318 |         "slides_video": "https://www.bilibili.com/video/BV1u64y1i75a",
319 |         "notebook_video": "",
320 |         "qa_video": ""
321 |     },
322 |     {
323 |         "title": "模型初始化和激活函数",
324 |         "day_break": false,
325 |         "slides": [
326 |             "part-01.pdf",
327 |             12
328 |         ],
329 |         "slides_video": "https://www.bilibili.com/video/BV1u64y1i75a?p=2",
330 |         "notebook_video": "",
331 |         "qa_video": ""
332 |     },
333 |     {
334 |         "title": "实战 Kaggle 比赛：预测房价",
335 |         "day_break": false,
336 |         "book": "chapter_multilayer-perceptrons/kaggle-house-price.html",
337 |         "slides": [
338 |             "part-0.pdf",
339 |             0
340 |         ],
341 |         "slides_video": "https://www.bilibili.com/video/BV1NK4y1P7Tu",
342 |         "notebook_video": "",
343 |         "qa_video": ""
344 |     },
345 |     {
346 |         "title": "**竞赛**{: .label  } 预测房价",
347 |         "day_break": false,
348 |         "book": "",
349 |         "slides": [
350 |             "part-01.pdf",
351 |             2
352 |         ],
353 |         "slides_video": "https://www.bilibili.com/video/BV1NK4y1P7Tu?p=2",
354 |         "notebook_video": "",
355 |         "qa_video": ""
356 |     }
357 | ]


--------------------------------------------------------------------------------
/_modules/part_0.md:
--------------------------------------------------------------------------------
  1 | ---
  2 | title: 深度学习基础
  3 | ---
  4 | 
  5 | 3月20日
  6 | 
  7 | : 课程安排
  8 |   : &nbsp; 
  9 |   : [<span class="iconfont icon-KeynoteOutline"></span>](assets/pdfs/part-0_1.pdf)
 10 |   :  &nbsp; 
 11 |   : [<span style="font-size:130%"  class="iconfont icon-bilibili-fill"></span>](https://www.bilibili.com/video/BV1oX4y137bC)
 12 | 
 13 | : 深度学习介绍
 14 |   : [<span class="iconfont icon-xiaoshuo-copy"></span>](https://zh-v2.d2l.ai/chapter_introduction/index.html)
 15 |   : [<span class="iconfont icon-KeynoteOutline"></span>](assets/pdfs/part-0_2.pdf)
 16 |   :  &nbsp; 
 17 |   : [<span style="font-size:130%"  class="iconfont icon-bilibili-fill"></span>](https://www.bilibili.com/video/BV1J54y187f9)
 18 | 
 19 | : 安装
 20 |   : [<span class="iconfont icon-xiaoshuo-copy"></span>](https://zh-v2.d2l.ai/chapter_installation/index.html)
 21 |   : [<span class="iconfont icon-KeynoteOutline"></span>](assets/pdfs/part-0_3.pdf)
 22 |   :  &nbsp; 
 23 |   : [<span style="font-size:130%"  class="iconfont icon-bilibili-fill"></span>](https://www.bilibili.com/video/BV18p4y1h7Dr)
 24 | 
 25 | : 数据操作
 26 |   : [<span class="iconfont icon-xiaoshuo-copy"></span>](https://zh-v2.d2l.ai/chapter_preliminaries/ndarray.html)
 27 |   : [<span class="iconfont icon-KeynoteOutline"></span>](assets/pdfs/part-0_4.pdf)
 28 |   : [<span class="iconfont icon-jupyter"></span>](assets/notebooks/chapter_preliminaries/ndarray.slides.html)
 29 |   : [<span style="font-size:130%"  class="iconfont icon-bilibili-fill"></span>](https://www.bilibili.com/video/BV1CV411Y7i4)
 30 | 
 31 | : 数据预处理
 32 |   : [<span class="iconfont icon-xiaoshuo-copy"></span>](https://zh-v2.d2l.ai/chapter_preliminaries/pandas.html)
 33 |   : &nbsp; 
 34 |   : [<span class="iconfont icon-jupyter"></span>](assets/notebooks/chapter_preliminaries/pandas.slides.html)
 35 |   : [<span style="font-size:130%"  class="iconfont icon-bilibili-fill"></span>](https://www.bilibili.com/video/BV1CV411Y7i4?p=3)
 36 | 
 37 | 
 38 | 3月21日
 39 | 
 40 | : 线性代数
 41 |   : [<span class="iconfont icon-xiaoshuo-copy"></span>](https://zh-v2.d2l.ai/chapter_preliminaries/linear-algebra.html)
 42 |   : [<span class="iconfont icon-KeynoteOutline"></span>](assets/pdfs/part-0_5.pdf)
 43 |   : [<span class="iconfont icon-jupyter"></span>](assets/notebooks/chapter_preliminaries/linear-algebra.slides.html)
 44 |   : [<span style="font-size:130%"  class="iconfont icon-bilibili-fill"></span>](https://www.bilibili.com/video/BV1eK4y1U7Qy)
 45 | 
 46 | : [补充] 按特定轴求和
 47 |   : &nbsp; 
 48 |   : &nbsp; 
 49 |   :  &nbsp; 
 50 |   : [<span style="font-size:130%"  class="iconfont icon-bilibili-fill"></span>](https://www.bilibili.com/video/BV1eK4y1U7Qy?p=3)
 51 | 
 52 | : 矩阵计算
 53 |   : [<span class="iconfont icon-xiaoshuo-copy"></span>](https://zh-v2.d2l.ai/chapter_preliminaries/calculus.html)
 54 |   : [<span class="iconfont icon-KeynoteOutline"></span>](assets/pdfs/part-0_6.pdf)
 55 |   :  &nbsp; 
 56 |   : [<span style="font-size:130%"  class="iconfont icon-bilibili-fill"></span>](https://www.bilibili.com/video/BV1eZ4y1w7PY)
 57 | 
 58 | : 自动求导
 59 |   : [<span class="iconfont icon-xiaoshuo-copy"></span>](https://zh-v2.d2l.ai/chapter_preliminaries/autograd.html)
 60 |   : [<span class="iconfont icon-KeynoteOutline"></span>](assets/pdfs/part-0_7.pdf)
 61 |   : [<span class="iconfont icon-jupyter"></span>](assets/notebooks/chapter_preliminaries/autograd.slides.html)
 62 |   : [<span style="font-size:130%"  class="iconfont icon-bilibili-fill"></span>](https://www.bilibili.com/video/BV1KA411N7Px)
 63 | 
 64 | 
 65 | 3月27日
 66 | 
 67 | : 线性回归
 68 |   : [<span class="iconfont icon-xiaoshuo-copy"></span>](https://zh-v2.d2l.ai/chapter_linear-networks/linear-regression.html)
 69 |   : [<span class="iconfont icon-KeynoteOutline"></span>](assets/pdfs/part-0_8.pdf)
 70 |   : [<span class="iconfont icon-jupyter"></span>](assets/notebooks/chapter_linear-networks/linear-regression.slides.html)
 71 |   : [<span style="font-size:130%"  class="iconfont icon-bilibili-fill"></span>](https://www.bilibili.com/video/BV1PX4y1g7KC)
 72 | 
 73 | : 基础优化方法
 74 |   : &nbsp; 
 75 |   : [<span class="iconfont icon-KeynoteOutline"></span>](assets/pdfs/part-0_9.pdf)
 76 |   :  &nbsp; 
 77 |   : [<span style="font-size:130%"  class="iconfont icon-bilibili-fill"></span>](https://www.bilibili.com/video/BV1PX4y1g7KC?p=2)
 78 | 
 79 | : 线性回归的从零开始实现
 80 |   : [<span class="iconfont icon-xiaoshuo-copy"></span>](https://zh-v2.d2l.ai/chapter_linear-networks/linear-regression-scratch.html)
 81 |   : &nbsp; 
 82 |   : [<span class="iconfont icon-jupyter"></span>](assets/notebooks/chapter_linear-networks/linear-regression-scratch.slides.html)
 83 |   : [<span style="font-size:130%"  class="iconfont icon-bilibili-fill"></span>](https://www.bilibili.com/video/BV1PX4y1g7KC?p=3)
 84 | 
 85 | : 线性回归的简洁实现
 86 |   : [<span class="iconfont icon-xiaoshuo-copy"></span>](https://zh-v2.d2l.ai/chapter_linear-networks/linear-regression-concise.html)
 87 |   : &nbsp; 
 88 |   : [<span class="iconfont icon-jupyter"></span>](assets/notebooks/chapter_linear-networks/linear-regression-concise.slides.html)
 89 |   : [<span style="font-size:130%"  class="iconfont icon-bilibili-fill"></span>](https://www.bilibili.com/video/BV1PX4y1g7KC?p=4)
 90 | 
 91 | 
 92 | 3月28日
 93 | 
 94 | : Softmax 回归
 95 |   : [<span class="iconfont icon-xiaoshuo-copy"></span>](https://zh-v2.d2l.ai/chapter_linear-networks/softmax-regression.html)
 96 |   : [<span class="iconfont icon-KeynoteOutline"></span>](assets/pdfs/part-0_10.pdf)
 97 |   :  &nbsp; 
 98 |   : [<span style="font-size:130%"  class="iconfont icon-bilibili-fill"></span>](https://www.bilibili.com/video/BV1K64y1Q7wu)
 99 | 
100 | : 损失函数
101 |   : &nbsp; 
102 |   : [<span class="iconfont icon-KeynoteOutline"></span>](assets/pdfs/part-0_11.pdf)
103 |   :  &nbsp; 
104 |   : [<span style="font-size:130%"  class="iconfont icon-bilibili-fill"></span>](https://www.bilibili.com/video/BV1K64y1Q7wu?p=2)
105 | 
106 | : 图像分类数据集
107 |   : [<span class="iconfont icon-xiaoshuo-copy"></span>](https://zh-v2.d2l.ai/chapter_linear-networks/image-classification-dataset.html)
108 |   : &nbsp; 
109 |   : [<span class="iconfont icon-jupyter"></span>](assets/notebooks/chapter_linear-networks/image-classification-dataset.slides.html)
110 |   : [<span style="font-size:130%"  class="iconfont icon-bilibili-fill"></span>](https://www.bilibili.com/video/BV1K64y1Q7wu?p=3)
111 | 
112 | : Softmax 回归的从零开始实现
113 |   : [<span class="iconfont icon-xiaoshuo-copy"></span>](https://zh-v2.d2l.ai/chapter_linear-networks/softmax-regression-scratch.html)
114 |   : &nbsp; 
115 |   : [<span class="iconfont icon-jupyter"></span>](assets/notebooks/chapter_linear-networks/softmax-regression-scratch.slides.html)
116 |   : [<span style="font-size:130%"  class="iconfont icon-bilibili-fill"></span>](https://www.bilibili.com/video/BV1K64y1Q7wu?p=4)
117 | 
118 | : Softmax 回归的简洁实现
119 |   : [<span class="iconfont icon-xiaoshuo-copy"></span>](https://zh-v2.d2l.ai/chapter_linear-networks/softmax-regression-concise.html)
120 |   : &nbsp; 
121 |   : [<span class="iconfont icon-jupyter"></span>](assets/notebooks/chapter_linear-networks/softmax-regression-concise.slides.html)
122 |   : [<span style="font-size:130%"  class="iconfont icon-bilibili-fill"></span>](https://www.bilibili.com/video/BV1K64y1Q7wu?p=5)
123 | 
124 | 
125 | 4月3日
126 | 
127 | : **休课**{: .label .label-green }
128 | 
129 | 4月4日
130 | 
131 | : **休课**{: .label .label-green }
132 | 
133 | 4月10日
134 | 
135 | : 感知机
136 |   : &nbsp; 
137 |   : [<span class="iconfont icon-KeynoteOutline"></span>](assets/pdfs/part-0_12.pdf)
138 |   :  &nbsp; 
139 |   : [<span style="font-size:130%"  class="iconfont icon-bilibili-fill"></span>](https://www.bilibili.com/video/BV1hh411U7gn)
140 | 
141 | : 多层感知机
142 |   : [<span class="iconfont icon-xiaoshuo-copy"></span>](https://zh-v2.d2l.ai/chapter_multilayer-perceptrons/mlp.html)
143 |   : [<span class="iconfont icon-KeynoteOutline"></span>](assets/pdfs/part-0_13.pdf)
144 |   : [<span class="iconfont icon-jupyter"></span>](assets/notebooks/chapter_multilayer-perceptrons/mlp.slides.html)
145 |   : [<span style="font-size:130%"  class="iconfont icon-bilibili-fill"></span>](https://www.bilibili.com/video/BV1hh411U7gn?p=2)
146 | 
147 | : 多层感知机的从零开始实现
148 |   : [<span class="iconfont icon-xiaoshuo-copy"></span>](https://zh-v2.d2l.ai/chapter_multilayer-perceptrons/mlp-scratch.html)
149 |   : &nbsp; 
150 |   : [<span class="iconfont icon-jupyter"></span>](assets/notebooks/chapter_multilayer-perceptrons/mlp-scratch.slides.html)
151 |   : [<span style="font-size:130%"  class="iconfont icon-bilibili-fill"></span>](https://www.bilibili.com/video/BV1hh411U7gn?p=3)
152 | 
153 | : 多层感知机的简洁实现
154 |   : [<span class="iconfont icon-xiaoshuo-copy"></span>](https://zh-v2.d2l.ai/chapter_multilayer-perceptrons/mlp-concise.html)
155 |   : &nbsp; 
156 |   : [<span class="iconfont icon-jupyter"></span>](assets/notebooks/chapter_multilayer-perceptrons/mlp-concise.slides.html)
157 |   : [<span style="font-size:130%"  class="iconfont icon-bilibili-fill"></span>](https://www.bilibili.com/video/BV1hh411U7gn?p=3)
158 | 
159 | 
160 | 4月11日
161 | 
162 | : 模型选择
163 |   : &nbsp; 
164 |   : [<span class="iconfont icon-KeynoteOutline"></span>](assets/pdfs/part-0_14.pdf)
165 |   :  &nbsp; 
166 |   : [<span style="font-size:130%"  class="iconfont icon-bilibili-fill"></span>](https://www.bilibili.com/video/BV1kX4y1g7jp)
167 | 
168 | : 欠拟合和过拟合
169 |   : [<span class="iconfont icon-xiaoshuo-copy"></span>](https://zh-v2.d2l.ai/chapter_multilayer-perceptrons/underfit-overfit.html)
170 |   : [<span class="iconfont icon-KeynoteOutline"></span>](assets/pdfs/part-0_15.pdf)
171 |   : [<span class="iconfont icon-jupyter"></span>](assets/notebooks/chapter_multilayer-perceptrons/underfit-overfit.slides.html)
172 |   : [<span style="font-size:130%"  class="iconfont icon-bilibili-fill"></span>](https://www.bilibili.com/video/BV1kX4y1g7jp?p=2)
173 | 
174 | 
175 | 4月17日
176 | 
177 | : 权重衰减
178 |   : [<span class="iconfont icon-xiaoshuo-copy"></span>](https://zh-v2.d2l.ai/chapter_multilayer-perceptrons/weight-decay.html)
179 |   : [<span class="iconfont icon-KeynoteOutline"></span>](assets/pdfs/part-0_16.pdf)
180 |   : [<span class="iconfont icon-jupyter"></span>](assets/notebooks/chapter_multilayer-perceptrons/weight-decay.slides.html)
181 |   : [<span style="font-size:130%"  class="iconfont icon-bilibili-fill"></span>](https://www.bilibili.com/video/BV1UK4y1o7dy)
182 | 
183 | : Dropout
184 |   : [<span class="iconfont icon-xiaoshuo-copy"></span>](https://zh-v2.d2l.ai/chapter_multilayer-perceptrons/dropout.html)
185 |   : [<span class="iconfont icon-KeynoteOutline"></span>](assets/pdfs/part-0_17.pdf)
186 |   : [<span class="iconfont icon-jupyter"></span>](assets/notebooks/chapter_multilayer-perceptrons/dropout.slides.html)
187 |   : [<span style="font-size:130%"  class="iconfont icon-bilibili-fill"></span>](https://www.bilibili.com/video/BV1Y5411c7aY)
188 | 
189 | 
190 | 4月18日
191 | 
192 | : 数值稳定性
193 |   : [<span class="iconfont icon-xiaoshuo-copy"></span>](https://zh-v2.d2l.ai/chapter_multilayer-perceptrons/numerical-stability-and-init.html)
194 |   : [<span class="iconfont icon-KeynoteOutline"></span>](assets/pdfs/part-0_18.pdf)
195 |   : [<span class="iconfont icon-jupyter"></span>](assets/notebooks/chapter_multilayer-perceptrons/numerical-stability-and-init.slides.html)
196 |   : [<span style="font-size:130%"  class="iconfont icon-bilibili-fill"></span>](https://www.bilibili.com/video/BV1u64y1i75a)
197 | 
198 | : 模型初始化和激活函数
199 |   : &nbsp; 
200 |   : [<span class="iconfont icon-KeynoteOutline"></span>](assets/pdfs/part-0_19.pdf)
201 |   :  &nbsp; 
202 |   : [<span style="font-size:130%"  class="iconfont icon-bilibili-fill"></span>](https://www.bilibili.com/video/BV1u64y1i75a?p=2)
203 | 
204 | : 实战 Kaggle 比赛：预测房价
205 |   : [<span class="iconfont icon-xiaoshuo-copy"></span>](https://zh-v2.d2l.ai/chapter_multilayer-perceptrons/kaggle-house-price.html)
206 |   : &nbsp; 
207 |   : [<span class="iconfont icon-jupyter"></span>](assets/notebooks/chapter_multilayer-perceptrons/kaggle-house-price.slides.html)
208 |   : [<span style="font-size:130%"  class="iconfont icon-bilibili-fill"></span>](https://www.bilibili.com/video/BV1NK4y1P7Tu)
209 | 
210 | : **竞赛**{: .label  } 预测房价
211 |   : &nbsp; 
212 |   : [<span class="iconfont icon-KeynoteOutline"></span>](assets/pdfs/part-0_20.pdf)
213 |   :  &nbsp; 
214 |   : [<span style="font-size:130%"  class="iconfont icon-bilibili-fill"></span>](https://www.bilibili.com/video/BV1NK4y1P7Tu?p=2)
215 | 
216 | 


--------------------------------------------------------------------------------
/_modules/part_1.json:
--------------------------------------------------------------------------------
  1 | [
  2 |     {
  3 |         "title": "模型构造",
  4 |         "day_break": true,
  5 |         "book": "chapter_deep-learning-computation/model-construction.html",
  6 |         "slides": [
  7 |             "part-0.pdf",
  8 |             0
  9 |         ],
 10 |         "slides_video": "https://www.bilibili.com/video/BV1AK4y1P7vs",
 11 |         "notebook_video": "",
 12 |         "qa_video": ""
 13 |     },
 14 |     {
 15 |         "title": "参数管理",
 16 |         "day_break": false,
 17 |         "book": "chapter_deep-learning-computation/parameters.html",
 18 |         "slides": [
 19 |             "part-0.pdf",
 20 |             0
 21 |         ],
 22 |         "slides_video": "https://www.bilibili.com/video/BV1AK4y1P7vs?p=2",
 23 |         "notebook_video": "",
 24 |         "qa_video": ""
 25 |     },
 26 |     {
 27 |         "title": "自定义层",
 28 |         "day_break": false,
 29 |         "book": "chapter_deep-learning-computation/custom-layer.html",
 30 |         "slides": [
 31 |             "part-0.pdf",
 32 |             0
 33 |         ],
 34 |         "slides_video": "https://www.bilibili.com/video/BV1AK4y1P7vs?p=3",
 35 |         "notebook_video": "",
 36 |         "qa_video": ""
 37 |     },
 38 |     {
 39 |         "title": "读写文件",
 40 |         "day_break": false,
 41 |         "book": "chapter_deep-learning-computation/read-write.html",
 42 |         "slides": [
 43 |             "part-0.pdf",
 44 |             0
 45 |         ],
 46 |         "slides_video": "https://www.bilibili.com/video/BV1AK4y1P7vs?p=4",
 47 |         "notebook_video": "",
 48 |         "qa_video": ""
 49 |     },
 50 |     {
 51 |         "title": "GPU",
 52 |         "day_break": false,
 53 |         "book": "chapter_deep-learning-computation/use-gpu.html",
 54 |         "slides": [
 55 |             "part-0.pdf",
 56 |             0
 57 |         ],
 58 |         "slides_video": "https://www.bilibili.com/video/BV1z5411c7C1",
 59 |         "notebook_video": "",
 60 |         "qa_video": ""
 61 |     },
 62 |     {
 63 |         "title": "预测房价竞赛总结",
 64 |         "day_break": true,
 65 |         "book": "",
 66 |         "slides": [
 67 |             "part-02.pdf",
 68 |             5
 69 |         ],
 70 |         "slides_video": "https://www.bilibili.com/video/BV15Q4y1o7vc",
 71 |         "notebook_video": "",
 72 |         "qa_video": ""
 73 |     },
 74 |     {
 75 |         "title": "从全连接层到卷积",
 76 |         "day_break": false,
 77 |         "book": "chapter_convolutional-neural-networks/why-conv.html",
 78 |         "slides": [
 79 |             "part-02.pdf",
 80 |             9
 81 |         ],
 82 |         "slides_video": "https://www.bilibili.com/video/BV1L64y1m7Nh",
 83 |         "notebook_video": "",
 84 |         "qa_video": ""
 85 |     },
 86 |     {
 87 |         "title": "图像卷积",
 88 |         "day_break": false,
 89 |         "book": "chapter_convolutional-neural-networks/conv-layer.html",
 90 |         "slides": [
 91 |             "part-02.pdf",
 92 |             7
 93 |         ],
 94 |         "slides_video": "https://www.bilibili.com/video/BV1L64y1m7Nh?p=2",
 95 |         "notebook_video": "",
 96 |         "qa_video": ""
 97 |     },
 98 |     {
 99 |         "title": "填充和步幅",
100 |         "day_break": true,
101 |         "book": "chapter_convolutional-neural-networks/padding-and-strides.html",
102 |         "slides": [
103 |             "part-02.pdf",
104 |             8
105 |         ],
106 |         "slides_video": "https://www.bilibili.com/video/BV1Th411U7UN",
107 |         "notebook_video": "",
108 |         "qa_video": ""
109 |     },
110 |     {
111 |         "title": "多输入多输出通道",
112 |         "day_break": false,
113 |         "book": "chapter_convolutional-neural-networks/channels.html",
114 |         "slides": [
115 |             "part-02.pdf",
116 |             10
117 |         ],
118 |         "slides_video": "https://www.bilibili.com/video/BV1MB4y1F7of",
119 |         "notebook_video": "",
120 |         "qa_video": ""
121 |     },
122 |     {
123 |         "title": "池化层",
124 |         "day_break": true,
125 |         "book": "chapter_convolutional-neural-networks/pooling.html",
126 |         "slides": [
127 |             "part-02.pdf",
128 |             7
129 |         ],
130 |         "slides_video": "https://www.bilibili.com/video/BV1EV411j7nX",
131 |         "notebook_video": "",
132 |         "qa_video": ""
133 |     },
134 |     {
135 |         "title": "卷积神经网络（LeNet）",
136 |         "day_break": false,
137 |         "book": "chapter_convolutional-neural-networks/lenet.html",
138 |         "slides": [
139 |             "part-02.pdf",
140 |             6
141 |         ],
142 |         "slides_video": "https://www.bilibili.com/video/BV1t44y1r7ct/",
143 |         "notebook_video": "",
144 |         "qa_video": ""
145 |     },
146 |     {
147 |         "title": "深度卷积神经网络（AlexNet）",
148 |         "day_break": true,
149 |         "book": "chapter_convolutional-modern/alexnet.html",
150 |         "slides": [
151 |             "part-02.pdf",
152 |             13
153 |         ],
154 |         "slides_video": "https://www.bilibili.com/video/BV1h54y1L7oe/",
155 |         "notebook_video": "",
156 |         "qa_video": ""
157 |     },
158 |     {
159 |         "title": "使用块的网络（VGG）",
160 |         "day_break": false,
161 |         "book": "chapter_convolutional-modern/vgg.html",
162 |         "slides": [
163 |             "part-02.pdf",
164 |             7
165 |         ],
166 |         "slides_video": "https://www.bilibili.com/video/BV1Ao4y117Pd/",
167 |         "notebook_video": "",
168 |         "qa_video": ""
169 |     },
170 |     {
171 |         "title": "网络中的网络（NiN）",
172 |         "day_break": true,
173 |         "book": "chapter_convolutional-modern/nin.html",
174 |         "slides": [
175 |             "part-02.pdf",
176 |             6
177 |         ],
178 |         "slides_video": "https://www.bilibili.com/video/BV1Uv411G71b/",
179 |         "notebook_video": "",
180 |         "qa_video": ""
181 |     },
182 |     {
183 |         "title": "含并行连结的网络（GoogLeNet）",
184 |         "day_break": false,
185 |         "book": "chapter_convolutional-modern/googlenet.html",
186 |         "slides": [
187 |             "part-02.pdf",
188 |             15
189 |         ],
190 |         "slides_video": "https://www.bilibili.com/video/BV1b5411g7Xo/",
191 |         "notebook_video": "",
192 |         "qa_video": ""
193 |     },
194 |     {
195 |         "title": "批量归一化",
196 |         "day_break": true,
197 |         "book": "chapter_convolutional-modern/batch-norm.html",
198 |         "slides": [
199 |             "part-02.pdf",
200 |             6
201 |         ],
202 |         "slides_video": "https://www.bilibili.com/video/BV1X44y1r77r/",
203 |         "notebook_video": "",
204 |         "qa_video": ""
205 |     },
206 |     {
207 |         "title": "残差网络（ResNet）",
208 |         "day_break": false,
209 |         "book": "chapter_convolutional-modern/resnet.html",
210 |         "slides": [
211 |             "part-02.pdf",
212 |             9
213 |         ],
214 |         "slides_video": "https://www.bilibili.com/video/BV1bV41177ap/",
215 |         "notebook_video": "",
216 |         "qa_video": ""
217 |     },
218 |     {
219 |         "title": "**竞赛**{: .label  } 图片分类",
220 |         "day_break": false,
221 |         "book": "",
222 |         "slides": [
223 |             "part-02.pdf",
224 |             2
225 |         ],
226 |         "slides_video": "https://www.bilibili.com/video/BV1z64y1o7iz/",
227 |         "notebook_video": "",
228 |         "qa_video": ""
229 |     }
230 | ]


--------------------------------------------------------------------------------
/_modules/part_1.md:
--------------------------------------------------------------------------------
  1 | ---
  2 | title: 卷积神经网络
  3 | ---
  4 | 
  5 | 4月24日
  6 | 
  7 | : 模型构造
  8 |   : [<span class="iconfont icon-xiaoshuo-copy"></span>](https://zh-v2.d2l.ai/chapter_deep-learning-computation/model-construction.html)
  9 |   : &nbsp; 
 10 |   : [<span class="iconfont icon-jupyter"></span>](assets/notebooks/chapter_deep-learning-computation/model-construction.slides.html)
 11 |   : [<span style="font-size:130%"  class="iconfont icon-bilibili-fill"></span>](https://www.bilibili.com/video/BV1AK4y1P7vs)
 12 | 
 13 | : 参数管理
 14 |   : [<span class="iconfont icon-xiaoshuo-copy"></span>](https://zh-v2.d2l.ai/chapter_deep-learning-computation/parameters.html)
 15 |   : &nbsp; 
 16 |   : [<span class="iconfont icon-jupyter"></span>](assets/notebooks/chapter_deep-learning-computation/parameters.slides.html)
 17 |   : [<span style="font-size:130%"  class="iconfont icon-bilibili-fill"></span>](https://www.bilibili.com/video/BV1AK4y1P7vs?p=2)
 18 | 
 19 | : 自定义层
 20 |   : [<span class="iconfont icon-xiaoshuo-copy"></span>](https://zh-v2.d2l.ai/chapter_deep-learning-computation/custom-layer.html)
 21 |   : &nbsp; 
 22 |   : [<span class="iconfont icon-jupyter"></span>](assets/notebooks/chapter_deep-learning-computation/custom-layer.slides.html)
 23 |   : [<span style="font-size:130%"  class="iconfont icon-bilibili-fill"></span>](https://www.bilibili.com/video/BV1AK4y1P7vs?p=3)
 24 | 
 25 | : 读写文件
 26 |   : [<span class="iconfont icon-xiaoshuo-copy"></span>](https://zh-v2.d2l.ai/chapter_deep-learning-computation/read-write.html)
 27 |   : &nbsp; 
 28 |   : [<span class="iconfont icon-jupyter"></span>](assets/notebooks/chapter_deep-learning-computation/read-write.slides.html)
 29 |   : [<span style="font-size:130%"  class="iconfont icon-bilibili-fill"></span>](https://www.bilibili.com/video/BV1AK4y1P7vs?p=4)
 30 | 
 31 | : GPU
 32 |   : [<span class="iconfont icon-xiaoshuo-copy"></span>](https://zh-v2.d2l.ai/chapter_deep-learning-computation/use-gpu.html)
 33 |   : &nbsp; 
 34 |   : [<span class="iconfont icon-jupyter"></span>](assets/notebooks/chapter_deep-learning-computation/use-gpu.slides.html)
 35 |   : [<span style="font-size:130%"  class="iconfont icon-bilibili-fill"></span>](https://www.bilibili.com/video/BV1z5411c7C1)
 36 | 
 37 | 
 38 | 4月25日
 39 | 
 40 | : **休课**{: .label .label-green }
 41 | 
 42 | 5月1日
 43 | 
 44 | : **休课**{: .label .label-green }
 45 | 
 46 | 5月2日
 47 | 
 48 | : **休课**{: .label .label-green }
 49 | 
 50 | 5月8日
 51 | 
 52 | : **休课**{: .label .label-green }
 53 | 
 54 | 5月9日
 55 | 
 56 | : **休课**{: .label .label-green }
 57 | 
 58 | 5月15日
 59 | 
 60 | : 预测房价竞赛总结
 61 |   : &nbsp; 
 62 |   : [<span class="iconfont icon-KeynoteOutline"></span>](assets/pdfs/part-1_1.pdf)
 63 |   :  &nbsp; 
 64 |   : [<span style="font-size:130%"  class="iconfont icon-bilibili-fill"></span>](https://www.bilibili.com/video/BV15Q4y1o7vc)
 65 | 
 66 | : 从全连接层到卷积
 67 |   : [<span class="iconfont icon-xiaoshuo-copy"></span>](https://zh-v2.d2l.ai/chapter_convolutional-neural-networks/why-conv.html)
 68 |   : [<span class="iconfont icon-KeynoteOutline"></span>](assets/pdfs/part-1_2.pdf)
 69 |   :  &nbsp; 
 70 |   : [<span style="font-size:130%"  class="iconfont icon-bilibili-fill"></span>](https://www.bilibili.com/video/BV1L64y1m7Nh)
 71 | 
 72 | : 图像卷积
 73 |   : [<span class="iconfont icon-xiaoshuo-copy"></span>](https://zh-v2.d2l.ai/chapter_convolutional-neural-networks/conv-layer.html)
 74 |   : [<span class="iconfont icon-KeynoteOutline"></span>](assets/pdfs/part-1_3.pdf)
 75 |   : [<span class="iconfont icon-jupyter"></span>](assets/notebooks/chapter_convolutional-neural-networks/conv-layer.slides.html)
 76 |   : [<span style="font-size:130%"  class="iconfont icon-bilibili-fill"></span>](https://www.bilibili.com/video/BV1L64y1m7Nh?p=2)
 77 | 
 78 | 
 79 | 5月16日
 80 | 
 81 | : 填充和步幅
 82 |   : [<span class="iconfont icon-xiaoshuo-copy"></span>](https://zh-v2.d2l.ai/chapter_convolutional-neural-networks/padding-and-strides.html)
 83 |   : [<span class="iconfont icon-KeynoteOutline"></span>](assets/pdfs/part-1_4.pdf)
 84 |   : [<span class="iconfont icon-jupyter"></span>](assets/notebooks/chapter_convolutional-neural-networks/padding-and-strides.slides.html)
 85 |   : [<span style="font-size:130%"  class="iconfont icon-bilibili-fill"></span>](https://www.bilibili.com/video/BV1Th411U7UN)
 86 | 
 87 | : 多输入多输出通道
 88 |   : [<span class="iconfont icon-xiaoshuo-copy"></span>](https://zh-v2.d2l.ai/chapter_convolutional-neural-networks/channels.html)
 89 |   : [<span class="iconfont icon-KeynoteOutline"></span>](assets/pdfs/part-1_5.pdf)
 90 |   : [<span class="iconfont icon-jupyter"></span>](assets/notebooks/chapter_convolutional-neural-networks/channels.slides.html)
 91 |   : [<span style="font-size:130%"  class="iconfont icon-bilibili-fill"></span>](https://www.bilibili.com/video/BV1MB4y1F7of)
 92 | 
 93 | 
 94 | 5月22日
 95 | 
 96 | : 池化层
 97 |   : [<span class="iconfont icon-xiaoshuo-copy"></span>](https://zh-v2.d2l.ai/chapter_convolutional-neural-networks/pooling.html)
 98 |   : [<span class="iconfont icon-KeynoteOutline"></span>](assets/pdfs/part-1_6.pdf)
 99 |   : [<span class="iconfont icon-jupyter"></span>](assets/notebooks/chapter_convolutional-neural-networks/pooling.slides.html)
100 |   : [<span style="font-size:130%"  class="iconfont icon-bilibili-fill"></span>](https://www.bilibili.com/video/BV1EV411j7nX)
101 | 
102 | : 卷积神经网络（LeNet）
103 |   : [<span class="iconfont icon-xiaoshuo-copy"></span>](https://zh-v2.d2l.ai/chapter_convolutional-neural-networks/lenet.html)
104 |   : [<span class="iconfont icon-KeynoteOutline"></span>](assets/pdfs/part-1_7.pdf)
105 |   : [<span class="iconfont icon-jupyter"></span>](assets/notebooks/chapter_convolutional-neural-networks/lenet.slides.html)
106 |   : [<span style="font-size:130%"  class="iconfont icon-bilibili-fill"></span>](https://www.bilibili.com/video/BV1t44y1r7ct/)
107 | 
108 | 
109 | 5月23日
110 | 
111 | : 深度卷积神经网络（AlexNet）
112 |   : [<span class="iconfont icon-xiaoshuo-copy"></span>](https://zh-v2.d2l.ai/chapter_convolutional-modern/alexnet.html)
113 |   : [<span class="iconfont icon-KeynoteOutline"></span>](assets/pdfs/part-1_8.pdf)
114 |   : [<span class="iconfont icon-jupyter"></span>](assets/notebooks/chapter_convolutional-modern/alexnet.slides.html)
115 |   : [<span style="font-size:130%"  class="iconfont icon-bilibili-fill"></span>](https://www.bilibili.com/video/BV1h54y1L7oe/)
116 | 
117 | : 使用块的网络（VGG）
118 |   : [<span class="iconfont icon-xiaoshuo-copy"></span>](https://zh-v2.d2l.ai/chapter_convolutional-modern/vgg.html)
119 |   : [<span class="iconfont icon-KeynoteOutline"></span>](assets/pdfs/part-1_9.pdf)
120 |   : [<span class="iconfont icon-jupyter"></span>](assets/notebooks/chapter_convolutional-modern/vgg.slides.html)
121 |   : [<span style="font-size:130%"  class="iconfont icon-bilibili-fill"></span>](https://www.bilibili.com/video/BV1Ao4y117Pd/)
122 | 
123 | 
124 | 5月29日
125 | 
126 | : 网络中的网络（NiN）
127 |   : [<span class="iconfont icon-xiaoshuo-copy"></span>](https://zh-v2.d2l.ai/chapter_convolutional-modern/nin.html)
128 |   : [<span class="iconfont icon-KeynoteOutline"></span>](assets/pdfs/part-1_10.pdf)
129 |   : [<span class="iconfont icon-jupyter"></span>](assets/notebooks/chapter_convolutional-modern/nin.slides.html)
130 |   : [<span style="font-size:130%"  class="iconfont icon-bilibili-fill"></span>](https://www.bilibili.com/video/BV1Uv411G71b/)
131 | 
132 | : 含并行连结的网络（GoogLeNet）
133 |   : [<span class="iconfont icon-xiaoshuo-copy"></span>](https://zh-v2.d2l.ai/chapter_convolutional-modern/googlenet.html)
134 |   : [<span class="iconfont icon-KeynoteOutline"></span>](assets/pdfs/part-1_11.pdf)
135 |   : [<span class="iconfont icon-jupyter"></span>](assets/notebooks/chapter_convolutional-modern/googlenet.slides.html)
136 |   : [<span style="font-size:130%"  class="iconfont icon-bilibili-fill"></span>](https://www.bilibili.com/video/BV1b5411g7Xo/)
137 | 
138 | 
139 | 5月30日
140 | 
141 | : 批量归一化
142 |   : [<span class="iconfont icon-xiaoshuo-copy"></span>](https://zh-v2.d2l.ai/chapter_convolutional-modern/batch-norm.html)
143 |   : [<span class="iconfont icon-KeynoteOutline"></span>](assets/pdfs/part-1_12.pdf)
144 |   : [<span class="iconfont icon-jupyter"></span>](assets/notebooks/chapter_convolutional-modern/batch-norm.slides.html)
145 |   : [<span style="font-size:130%"  class="iconfont icon-bilibili-fill"></span>](https://www.bilibili.com/video/BV1X44y1r77r/)
146 | 
147 | : 残差网络（ResNet）
148 |   : [<span class="iconfont icon-xiaoshuo-copy"></span>](https://zh-v2.d2l.ai/chapter_convolutional-modern/resnet.html)
149 |   : [<span class="iconfont icon-KeynoteOutline"></span>](assets/pdfs/part-1_13.pdf)
150 |   : [<span class="iconfont icon-jupyter"></span>](assets/notebooks/chapter_convolutional-modern/resnet.slides.html)
151 |   : [<span style="font-size:130%"  class="iconfont icon-bilibili-fill"></span>](https://www.bilibili.com/video/BV1bV41177ap/)
152 | 
153 | : **竞赛**{: .label  } 图片分类
154 |   : &nbsp; 
155 |   : [<span class="iconfont icon-KeynoteOutline"></span>](assets/pdfs/part-1_14.pdf)
156 |   :  &nbsp; 
157 |   : [<span style="font-size:130%"  class="iconfont icon-bilibili-fill"></span>](https://www.bilibili.com/video/BV1z64y1o7iz/)
158 | 
159 | 


--------------------------------------------------------------------------------
/_modules/part_2.json:
--------------------------------------------------------------------------------
  1 | [
  2 |     {
  3 |         "title": "硬件：CPU和GPU",
  4 |         "day_break": true,
  5 |         "book": "chapter_computational-performance/hardware.html",
  6 |         "slides": [
  7 |             "part-03.pdf",
  8 |             14
  9 |         ],
 10 |         "slides_video": "https://www.bilibili.com/video/BV1TU4y1j7Wd/",
 11 |         "notebook_video": "",
 12 |         "qa_video": ""
 13 |     },
 14 |     {
 15 |         "title": "更多的专有硬件",
 16 |         "day_break": true,
 17 |         "book": "",
 18 |         "slides": [
 19 |             "part-03.pdf",
 20 |             16
 21 |         ],
 22 |         "slides_video": "https://www.bilibili.com/video/BV1VV41147PC/",
 23 |         "notebook_video": "",
 24 |         "qa_video": ""
 25 |     },
 26 |     {
 27 |         "title": "多GPU训练",
 28 |         "day_break": false,
 29 |         "book": "chapter_computational-performance/multiple-gpus.html",
 30 |         "slides": [
 31 |             "part-03.pdf",
 32 |             5
 33 |         ],
 34 |         "slides_video": "https://www.bilibili.com/video/BV1vU4y1V7rd/",
 35 |         "notebook_video": "",
 36 |         "qa_video": ""
 37 |     },
 38 |     {
 39 |         "title": "多GPU训练的实现",
 40 |         "day_break": true,
 41 |         "book": "chapter_computational-performance/multiple-gpus-concise.html",
 42 |         "slides": [
 43 |             "part-0.pdf",
 44 |             0
 45 |         ],
 46 |         "slides_video": "https://www.bilibili.com/video/BV1MQ4y1R7Qg",
 47 |         "notebook_video": "",
 48 |         "qa_video": ""
 49 |     },
 50 |     {
 51 |         "title": "分布式训练",
 52 |         "day_break": false,
 53 |         "book": "chapter_computational-performance/parameterserver.html",
 54 |         "slides": [
 55 |             "part-03.pdf",
 56 |             16
 57 |         ],
 58 |         "slides_video": "https://www.bilibili.com/video/BV1jU4y1G7iu",
 59 |         "notebook_video": "",
 60 |         "qa_video": ""
 61 |     },
 62 |     {
 63 |         "title": "图像增广",
 64 |         "day_break": true,
 65 |         "book": "chapter_computer-vision/image-augmentation.html",
 66 |         "slides": [
 67 |             "part-03.pdf",
 68 |             9
 69 |         ],
 70 |         "slides_video": "https://www.bilibili.com/video/BV17y4y1g76q",
 71 |         "notebook_video": "",
 72 |         "qa_video": ""
 73 |     },
 74 |     {
 75 |         "title": "微调",
 76 |         "day_break": false,
 77 |         "book": "chapter_computer-vision/fine-tuning.html",
 78 |         "slides": [
 79 |             "part-03.pdf",
 80 |             9
 81 |         ],
 82 |         "slides_video": "https://www.bilibili.com/video/BV1Sb4y1d7CR",
 83 |         "notebook_video": "",
 84 |         "qa_video": ""
 85 |     },
 86 |     {
 87 |         "title": "实战 Kaggle 比赛：图像分类（CIFAR-10）",
 88 |         "day_break": true,
 89 |         "book": "chapter_computer-vision/kaggle-cifar10.html",
 90 |         "slides": [
 91 |             "part-0.pdf",
 92 |             0
 93 |         ],
 94 |         "slides_video": "https://www.bilibili.com/video/BV1Gy4y1M7Cu",
 95 |         "notebook_video": "",
 96 |         "qa_video": ""
 97 |     },
 98 |     {
 99 |         "title": "实战 Kaggle 比赛：狗的品种识别（ImageNet Dogs）",
100 |         "day_break": false,
101 |         "book": "chapter_computer-vision/kaggle-dog.html",
102 |         "slides": [
103 |             "part-0.pdf",
104 |             0
105 |         ],
106 |         "slides_video": "https://www.bilibili.com/video/BV1j5411T7wx",
107 |         "notebook_video": "",
108 |         "qa_video": ""
109 |     },
110 |     {
111 |         "title": "物体检测",
112 |         "day_break": true,
113 |         "book": "",
114 |         "slides": [
115 |             "part-03.pdf",
116 |             6
117 |         ],
118 |         "slides_video": "https://www.bilibili.com/video/BV1Lh411Y7LX",
119 |         "notebook_video": "",
120 |         "qa_video": ""
121 |     },
122 |     {
123 |         "title": "边缘框实现",
124 |         "day_break": false,
125 |         "book": "chapter_computer-vision/bounding-box.html",
126 |         "slides": [
127 |             "part-03.pdf",
128 |             0
129 |         ],
130 |         "slides_video": "https://www.bilibili.com/video/BV1Lh411Y7LX?p=2",
131 |         "notebook_video": "",
132 |         "qa_video": ""
133 |     },
134 |     
135 |     {
136 |         "title": "物体检测数据集",
137 |         "day_break": false,
138 |         "book": "chapter_computer-vision/object-detection-dataset.html",
139 |         "slides": [
140 |             "part-0.pdf",
141 |             0
142 |         ],
143 |         "slides_video": "https://www.bilibili.com/video/BV1Lh411Y7LX?p=3",
144 |         "notebook_video": "",
145 |         "qa_video": ""
146 |     },
147 |     {
148 |         "title": "锚框",
149 |         "day_break": false,
150 |         "book": "chapter_computer-vision/anchor.html",
151 |         "slides": [
152 |             "part-03.pdf",
153 |             7
154 |         ],
155 |         "slides_video": "https://www.bilibili.com/video/BV1aB4y1K7za",
156 |         "notebook_video": "",
157 |         "qa_video": ""
158 |     },
159 |     {
160 |         "title": "**竞赛**{: .label  } 树叶分类竞赛总结",
161 |         "day_break": false,
162 |         "book": "",
163 |         "slides": [
164 |             "part-03.pdf",
165 |             8
166 |         ],
167 |         "slides_video": "https://www.bilibili.com/video/BV1by4y1K7SE",
168 |         "notebook_video": "",
169 |         "qa_video": ""
170 |     },
171 |     {
172 |         "title": "区域卷积神经网络（R-CNNs）",
173 |         "day_break": true,
174 |         "book": "chapter_computer-vision/rcnn.html",
175 |         "slides": [
176 |             "part-03.pdf",
177 |             8
178 |         ],
179 |         "slides_video": "https://www.bilibili.com/video/BV1Db4y1C71g",
180 |         "notebook_video": "",
181 |         "qa_video": ""
182 |     },
183 |     {
184 |         "title": "单发多框检测（SSD）",
185 |         "day_break": false,
186 |         "book": "",
187 |         "slides": [
188 |             "part-03.pdf",
189 |             5
190 |         ],
191 |         "slides_video": "https://www.bilibili.com/video/BV1Db4y1C71g",
192 |         "notebook_video": "",
193 |         "qa_video": ""
194 |     },    
195 |     {
196 |         "title": "你只看一次（YOLO）",
197 |         "day_break": false,
198 |         "book": "",
199 |         "slides": [
200 |             "part-03.pdf",
201 |             3
202 |         ],
203 |         "slides_video": "https://www.bilibili.com/video/BV1Db4y1C71g",
204 |         "notebook_video": "",
205 |         "qa_video": ""
206 |     },
207 |     {
208 |         "title": "多尺度物体检测实现",
209 |         "day_break": true,
210 |         "book": "chapter_computer-vision/multiscale-object-detection.html",
211 |         "slides": [
212 |             "part-0.pdf",
213 |             0
214 |         ],
215 |         "slides_video": "https://www.bilibili.com/video/BV1ZX4y1c7Sw?p=1",
216 |         "notebook_video": "",
217 |         "qa_video": ""
218 |     },
219 |     {
220 |         "title": "SSD 实现",
221 |         "day_break": false,
222 |         "book": "chapter_computer-vision/ssd.html",
223 |         "slides": [
224 |             "part-0.pdf",
225 |             0
226 |         ],
227 |         "slides_video": "https://www.bilibili.com/video/BV1ZX4y1c7Sw?p=2",
228 |         "notebook_video": "",
229 |         "qa_video": ""
230 |     },
231 |     {
232 |         "title": "语义分割",
233 |         "day_break": true,
234 |         "book": "",
235 |         "slides": [
236 |             "part-03.pdf",
237 |             5
238 |         ],
239 |         "slides_video": "https://www.bilibili.com/video/BV1BK4y1M7Rd",
240 |         "notebook_video": "",
241 |         "qa_video": ""
242 |     },
243 |     {
244 |         "title": "语义分割数据集",
245 |         "day_break": false,
246 |         "book": "chapter_computer-vision/semantic-segmentation-and-dataset.html",
247 |         "slides": [
248 |             "part-0.pdf",
249 |             0
250 |         ],
251 |         "slides_video": "https://www.bilibili.com/video/BV1BK4y1M7Rd?p=2",
252 |         "notebook_video": "",
253 |         "qa_video": ""
254 |     },
255 |     {
256 |         "title": "转置卷积",
257 |         "day_break": false,
258 |         "book": "chapter_computer-vision/transposed-conv.html",
259 |         "slides": [
260 |             "part-03.pdf",
261 |             3
262 |         ],
263 |         "slides_video": "https://www.bilibili.com/video/BV17o4y1X7Jn/",
264 |         "notebook_video": "",
265 |         "qa_video": ""
266 |     },
267 |     {
268 |         "title": "转置卷积是一种卷积",
269 |         "day_break": false,
270 |         "book": "chapter_computer-vision/transposed-conv.html",
271 |         "slides": [
272 |             "part-03.pdf",
273 |             8
274 |         ],
275 |         "slides_video": "https://www.bilibili.com/video/BV1CM4y1K7r7/",
276 |         "notebook_video": "",
277 |         "qa_video": ""
278 |     },
279 |     {
280 |         "title": "全连接卷积神经网络（FCN）",
281 |         "day_break": true,
282 |         "book": "chapter_computer-vision/fcn.html",
283 |         "slides": [
284 |             "part-03.pdf",
285 |             2
286 |         ],
287 |         "slides_video": "https://www.bilibili.com/video/BV1af4y1L7Zu/",
288 |         "notebook_video": "",
289 |         "qa_video": ""
290 |     },
291 |     {
292 |         "title": "样式迁移",
293 |         "day_break": false,
294 |         "book": "chapter_computer-vision/neural-style.html",
295 |         "slides": [
296 |             "part-03.pdf",
297 |             3
298 |         ],
299 |         "slides_video": "https://www.bilibili.com/video/BV1Eh41167GN/",
300 |         "notebook_video": "",
301 |         "qa_video": ""
302 |     },
303 |     {
304 |         "title": "**竞赛**{: .label  } 目标检测",
305 |         "day_break": false,
306 |         "book": "",
307 |         "slides": [
308 |             "part-03.pdf",
309 |             3
310 |         ],
311 |         "slides_video": "https://www.bilibili.com/video/BV1F64y1x7xP/",
312 |         "notebook_video": "",
313 |         "qa_video": ""
314 |     }
315 | ]


--------------------------------------------------------------------------------
/_modules/part_2.md:
--------------------------------------------------------------------------------
  1 | ---
  2 | title: 计算机视觉
  3 | ---
  4 | 
  5 | 6月5日
  6 | 
  7 | : 硬件：CPU和GPU
  8 |   : [<span class="iconfont icon-xiaoshuo-copy"></span>](https://zh-v2.d2l.ai/chapter_computational-performance/hardware.html)
  9 |   : [<span class="iconfont icon-KeynoteOutline"></span>](assets/pdfs/part-2_1.pdf)
 10 |   :  &nbsp; 
 11 |   : [<span style="font-size:130%"  class="iconfont icon-bilibili-fill"></span>](https://www.bilibili.com/video/BV1TU4y1j7Wd/)
 12 | 
 13 | 
 14 | 6月6日
 15 | 
 16 | : 更多的专有硬件
 17 |   : &nbsp; 
 18 |   : [<span class="iconfont icon-KeynoteOutline"></span>](assets/pdfs/part-2_2.pdf)
 19 |   :  &nbsp; 
 20 |   : [<span style="font-size:130%"  class="iconfont icon-bilibili-fill"></span>](https://www.bilibili.com/video/BV1VV41147PC/)
 21 | 
 22 | : 多GPU训练
 23 |   : [<span class="iconfont icon-xiaoshuo-copy"></span>](https://zh-v2.d2l.ai/chapter_computational-performance/multiple-gpus.html)
 24 |   : [<span class="iconfont icon-KeynoteOutline"></span>](assets/pdfs/part-2_3.pdf)
 25 |   : [<span class="iconfont icon-jupyter"></span>](assets/notebooks/chapter_computational-performance/multiple-gpus.slides.html)
 26 |   : [<span style="font-size:130%"  class="iconfont icon-bilibili-fill"></span>](https://www.bilibili.com/video/BV1vU4y1V7rd/)
 27 | 
 28 | 
 29 | 6月12日
 30 | 
 31 | : **休课**{: .label .label-green }
 32 | 
 33 | 6月13日
 34 | 
 35 | : **休课**{: .label .label-green }
 36 | 
 37 | 6月19日
 38 | 
 39 | : 多GPU训练的实现
 40 |   : [<span class="iconfont icon-xiaoshuo-copy"></span>](https://zh-v2.d2l.ai/chapter_computational-performance/multiple-gpus-concise.html)
 41 |   : &nbsp; 
 42 |   : [<span class="iconfont icon-jupyter"></span>](assets/notebooks/chapter_computational-performance/multiple-gpus-concise.slides.html)
 43 |   : [<span style="font-size:130%"  class="iconfont icon-bilibili-fill"></span>](https://www.bilibili.com/video/BV1MQ4y1R7Qg)
 44 | 
 45 | : 分布式训练
 46 |   : [<span class="iconfont icon-xiaoshuo-copy"></span>](https://zh-v2.d2l.ai/chapter_computational-performance/parameterserver.html)
 47 |   : [<span class="iconfont icon-KeynoteOutline"></span>](assets/pdfs/part-2_4.pdf)
 48 |   :  &nbsp; 
 49 |   : [<span style="font-size:130%"  class="iconfont icon-bilibili-fill"></span>](https://www.bilibili.com/video/BV1jU4y1G7iu)
 50 | 
 51 | 
 52 | 6月20日
 53 | 
 54 | : 图像增广
 55 |   : [<span class="iconfont icon-xiaoshuo-copy"></span>](https://zh-v2.d2l.ai/chapter_computer-vision/image-augmentation.html)
 56 |   : [<span class="iconfont icon-KeynoteOutline"></span>](assets/pdfs/part-2_5.pdf)
 57 |   : [<span class="iconfont icon-jupyter"></span>](assets/notebooks/chapter_computer-vision/image-augmentation.slides.html)
 58 |   : [<span style="font-size:130%"  class="iconfont icon-bilibili-fill"></span>](https://www.bilibili.com/video/BV17y4y1g76q)
 59 | 
 60 | : 微调
 61 |   : [<span class="iconfont icon-xiaoshuo-copy"></span>](https://zh-v2.d2l.ai/chapter_computer-vision/fine-tuning.html)
 62 |   : [<span class="iconfont icon-KeynoteOutline"></span>](assets/pdfs/part-2_6.pdf)
 63 |   : [<span class="iconfont icon-jupyter"></span>](assets/notebooks/chapter_computer-vision/fine-tuning.slides.html)
 64 |   : [<span style="font-size:130%"  class="iconfont icon-bilibili-fill"></span>](https://www.bilibili.com/video/BV1Sb4y1d7CR)
 65 | 
 66 | 
 67 | 6月26日
 68 | 
 69 | : 实战 Kaggle 比赛：图像分类（CIFAR-10）
 70 |   : [<span class="iconfont icon-xiaoshuo-copy"></span>](https://zh-v2.d2l.ai/chapter_computer-vision/kaggle-cifar10.html)
 71 |   : &nbsp; 
 72 |   : [<span class="iconfont icon-jupyter"></span>](assets/notebooks/chapter_computer-vision/kaggle-cifar10.slides.html)
 73 |   : [<span style="font-size:130%"  class="iconfont icon-bilibili-fill"></span>](https://www.bilibili.com/video/BV1Gy4y1M7Cu)
 74 | 
 75 | : 实战 Kaggle 比赛：狗的品种识别（ImageNet Dogs）
 76 |   : [<span class="iconfont icon-xiaoshuo-copy"></span>](https://zh-v2.d2l.ai/chapter_computer-vision/kaggle-dog.html)
 77 |   : &nbsp; 
 78 |   : [<span class="iconfont icon-jupyter"></span>](assets/notebooks/chapter_computer-vision/kaggle-dog.slides.html)
 79 |   : [<span style="font-size:130%"  class="iconfont icon-bilibili-fill"></span>](https://www.bilibili.com/video/BV1j5411T7wx)
 80 | 
 81 | 
 82 | 6月27日
 83 | 
 84 | : 物体检测
 85 |   : &nbsp; 
 86 |   : [<span class="iconfont icon-KeynoteOutline"></span>](assets/pdfs/part-2_7.pdf)
 87 |   :  &nbsp; 
 88 |   : [<span style="font-size:130%"  class="iconfont icon-bilibili-fill"></span>](https://www.bilibili.com/video/BV1Lh411Y7LX)
 89 | 
 90 | : 边缘框实现
 91 |   : [<span class="iconfont icon-xiaoshuo-copy"></span>](https://zh-v2.d2l.ai/chapter_computer-vision/bounding-box.html)
 92 |   : &nbsp; 
 93 |   : [<span class="iconfont icon-jupyter"></span>](assets/notebooks/chapter_computer-vision/bounding-box.slides.html)
 94 |   : [<span style="font-size:130%"  class="iconfont icon-bilibili-fill"></span>](https://www.bilibili.com/video/BV1Lh411Y7LX?p=2)
 95 | 
 96 | : 物体检测数据集
 97 |   : [<span class="iconfont icon-xiaoshuo-copy"></span>](https://zh-v2.d2l.ai/chapter_computer-vision/object-detection-dataset.html)
 98 |   : &nbsp; 
 99 |   : [<span class="iconfont icon-jupyter"></span>](assets/notebooks/chapter_computer-vision/object-detection-dataset.slides.html)
100 |   : [<span style="font-size:130%"  class="iconfont icon-bilibili-fill"></span>](https://www.bilibili.com/video/BV1Lh411Y7LX?p=3)
101 | 
102 | : 锚框
103 |   : [<span class="iconfont icon-xiaoshuo-copy"></span>](https://zh-v2.d2l.ai/chapter_computer-vision/anchor.html)
104 |   : [<span class="iconfont icon-KeynoteOutline"></span>](assets/pdfs/part-2_8.pdf)
105 |   : [<span class="iconfont icon-jupyter"></span>](assets/notebooks/chapter_computer-vision/anchor.slides.html)
106 |   : [<span style="font-size:130%"  class="iconfont icon-bilibili-fill"></span>](https://www.bilibili.com/video/BV1aB4y1K7za)
107 | 
108 | : **竞赛**{: .label  } 树叶分类竞赛总结
109 |   : &nbsp; 
110 |   : [<span class="iconfont icon-KeynoteOutline"></span>](assets/pdfs/part-2_9.pdf)
111 |   :  &nbsp; 
112 |   : [<span style="font-size:130%"  class="iconfont icon-bilibili-fill"></span>](https://www.bilibili.com/video/BV1by4y1K7SE)
113 | 
114 | 
115 | 7月3日
116 | 
117 | : 区域卷积神经网络（R-CNNs）
118 |   : [<span class="iconfont icon-xiaoshuo-copy"></span>](https://zh-v2.d2l.ai/chapter_computer-vision/rcnn.html)
119 |   : [<span class="iconfont icon-KeynoteOutline"></span>](assets/pdfs/part-2_10.pdf)
120 |   :  &nbsp; 
121 |   : [<span style="font-size:130%"  class="iconfont icon-bilibili-fill"></span>](https://www.bilibili.com/video/BV1Db4y1C71g)
122 | 
123 | : 单发多框检测（SSD）
124 |   : &nbsp; 
125 |   : [<span class="iconfont icon-KeynoteOutline"></span>](assets/pdfs/part-2_11.pdf)
126 |   :  &nbsp; 
127 |   : [<span style="font-size:130%"  class="iconfont icon-bilibili-fill"></span>](https://www.bilibili.com/video/BV1Db4y1C71g)
128 | 
129 | : 你只看一次（YOLO）
130 |   : &nbsp; 
131 |   : [<span class="iconfont icon-KeynoteOutline"></span>](assets/pdfs/part-2_12.pdf)
132 |   :  &nbsp; 
133 |   : [<span style="font-size:130%"  class="iconfont icon-bilibili-fill"></span>](https://www.bilibili.com/video/BV1Db4y1C71g)
134 | 
135 | 
136 | 7月4日
137 | 
138 | : 多尺度物体检测实现
139 |   : [<span class="iconfont icon-xiaoshuo-copy"></span>](https://zh-v2.d2l.ai/chapter_computer-vision/multiscale-object-detection.html)
140 |   : &nbsp; 
141 |   : [<span class="iconfont icon-jupyter"></span>](assets/notebooks/chapter_computer-vision/multiscale-object-detection.slides.html)
142 |   : [<span style="font-size:130%"  class="iconfont icon-bilibili-fill"></span>](https://www.bilibili.com/video/BV1ZX4y1c7Sw?p=1)
143 | 
144 | : SSD 实现
145 |   : [<span class="iconfont icon-xiaoshuo-copy"></span>](https://zh-v2.d2l.ai/chapter_computer-vision/ssd.html)
146 |   : &nbsp; 
147 |   : [<span class="iconfont icon-jupyter"></span>](assets/notebooks/chapter_computer-vision/ssd.slides.html)
148 |   : [<span style="font-size:130%"  class="iconfont icon-bilibili-fill"></span>](https://www.bilibili.com/video/BV1ZX4y1c7Sw?p=2)
149 | 
150 | 
151 | 7月10日
152 | 
153 | : 语义分割
154 |   : &nbsp; 
155 |   : [<span class="iconfont icon-KeynoteOutline"></span>](assets/pdfs/part-2_13.pdf)
156 |   :  &nbsp; 
157 |   : [<span style="font-size:130%"  class="iconfont icon-bilibili-fill"></span>](https://www.bilibili.com/video/BV1BK4y1M7Rd)
158 | 
159 | : 语义分割数据集
160 |   : [<span class="iconfont icon-xiaoshuo-copy"></span>](https://zh-v2.d2l.ai/chapter_computer-vision/semantic-segmentation-and-dataset.html)
161 |   : &nbsp; 
162 |   : [<span class="iconfont icon-jupyter"></span>](assets/notebooks/chapter_computer-vision/semantic-segmentation-and-dataset.slides.html)
163 |   : [<span style="font-size:130%"  class="iconfont icon-bilibili-fill"></span>](https://www.bilibili.com/video/BV1BK4y1M7Rd?p=2)
164 | 
165 | : 转置卷积
166 |   : [<span class="iconfont icon-xiaoshuo-copy"></span>](https://zh-v2.d2l.ai/chapter_computer-vision/transposed-conv.html)
167 |   : [<span class="iconfont icon-KeynoteOutline"></span>](assets/pdfs/part-2_14.pdf)
168 |   : [<span class="iconfont icon-jupyter"></span>](assets/notebooks/chapter_computer-vision/transposed-conv.slides.html)
169 |   : [<span style="font-size:130%"  class="iconfont icon-bilibili-fill"></span>](https://www.bilibili.com/video/BV17o4y1X7Jn/)
170 | 
171 | : 转置卷积是一种卷积
172 |   : [<span class="iconfont icon-xiaoshuo-copy"></span>](https://zh-v2.d2l.ai/chapter_computer-vision/transposed-conv.html)
173 |   : [<span class="iconfont icon-KeynoteOutline"></span>](assets/pdfs/part-2_15.pdf)
174 |   : [<span class="iconfont icon-jupyter"></span>](assets/notebooks/chapter_computer-vision/transposed-conv.slides.html)
175 |   : [<span style="font-size:130%"  class="iconfont icon-bilibili-fill"></span>](https://www.bilibili.com/video/BV1CM4y1K7r7/)
176 | 
177 | 
178 | 7月11日
179 | 
180 | : 全连接卷积神经网络（FCN）
181 |   : [<span class="iconfont icon-xiaoshuo-copy"></span>](https://zh-v2.d2l.ai/chapter_computer-vision/fcn.html)
182 |   : [<span class="iconfont icon-KeynoteOutline"></span>](assets/pdfs/part-2_16.pdf)
183 |   : [<span class="iconfont icon-jupyter"></span>](assets/notebooks/chapter_computer-vision/fcn.slides.html)
184 |   : [<span style="font-size:130%"  class="iconfont icon-bilibili-fill"></span>](https://www.bilibili.com/video/BV1af4y1L7Zu/)
185 | 
186 | : 样式迁移
187 |   : [<span class="iconfont icon-xiaoshuo-copy"></span>](https://zh-v2.d2l.ai/chapter_computer-vision/neural-style.html)
188 |   : [<span class="iconfont icon-KeynoteOutline"></span>](assets/pdfs/part-2_17.pdf)
189 |   : [<span class="iconfont icon-jupyter"></span>](assets/notebooks/chapter_computer-vision/neural-style.slides.html)
190 |   : [<span style="font-size:130%"  class="iconfont icon-bilibili-fill"></span>](https://www.bilibili.com/video/BV1Eh41167GN/)
191 | 
192 | : **竞赛**{: .label  } 目标检测
193 |   : &nbsp; 
194 |   : [<span class="iconfont icon-KeynoteOutline"></span>](assets/pdfs/part-2_18.pdf)
195 |   :  &nbsp; 
196 |   : [<span style="font-size:130%"  class="iconfont icon-bilibili-fill"></span>](https://www.bilibili.com/video/BV1F64y1x7xP/)
197 | 
198 | 


--------------------------------------------------------------------------------
/_modules/part_3.json:
--------------------------------------------------------------------------------
  1 | [
  2 |     {
  3 |         "title":"序列模型",
  4 |         "day_break":true,
  5 |         "book":"chapter_recurrent-neural-networks/sequence.html",
  6 |         "slides":["part-04.pdf",9],
  7 |         "slides_video":"https://www.bilibili.com/video/BV1L44y1m768/",
  8 |         "notebook_video":"",
  9 |         "qa_video":""
 10 |     },
 11 |     {
 12 |         "title":"文本预处理",
 13 |         "day_break":false,
 14 |         "book":"chapter_recurrent-neural-networks/text-preprocessing.html",
 15 |         "slides":["part-0.pdf",0],
 16 |         "slides_video":"https://www.bilibili.com/video/BV1Fo4y1Q79L/",
 17 |         "notebook_video":"",
 18 |         "qa_video":""
 19 |     },
 20 |     {
 21 |         "title":"语言模型和数据集",
 22 |         "day_break":true,
 23 |         "book":"chapter_recurrent-neural-networks/language-models-and-dataset.html",
 24 |         "slides":["part-04.pdf",5],
 25 |         "slides_video":"https://www.bilibili.com/video/BV1ZX4y1F7K3/",
 26 |         "notebook_video":"",
 27 |         "qa_video":""
 28 |     },
 29 |     {
 30 |         "title":"循环神经网络",
 31 |         "day_break":false,
 32 |         "book":"chapter_recurrent-neural-networks/rnn.html",
 33 |         "slides":["part-04.pdf",8],
 34 |         "slides_video":"https://www.bilibili.com/video/BV1D64y1z7CA/",
 35 |         "notebook_video":"",
 36 |         "qa_video":""
 37 |     },
 38 |     {
 39 |         "title":"循环神经网络的从零开始实现",
 40 |         "day_break":true,
 41 |         "book":"chapter_recurrent-neural-networks/rnn-scratch.html",
 42 |         "slides":["part-0.pdf",0],
 43 |         "slides_video":"https://www.bilibili.com/video/BV1kq4y1H7sw/",
 44 |         "notebook_video":"",
 45 |         "qa_video":""
 46 |     },
 47 |     {
 48 |         "title":"循环神经网络的简洁实现",
 49 |         "day_break":false,
 50 |         "book":"chapter_recurrent-neural-networks/rnn-concise.html",
 51 |         "slides":["part-0.pdf",0],
 52 |         "slides_video":"https://www.bilibili.com/video/BV1kq4y1H7sw?p=2",
 53 |         "notebook_video":"",
 54 |         "qa_video":""
 55 |     },
 56 |     {
 57 |         "title":"门控循环单元（GRU）",
 58 |         "day_break":true,
 59 |         "book":"chapter_recurrent-modern/gru.html",
 60 |         "slides":["part-04.pdf",6],
 61 |         "slides_video":"https://www.bilibili.com/video/BV1mf4y157N2/",
 62 |         "notebook_video":"",
 63 |         "qa_video":""
 64 |     },
 65 |     {
 66 |         "title":"长短期记忆网络（LSTM）",
 67 |         "day_break":false,
 68 |         "book":"chapter_recurrent-modern/lstm.html",
 69 |         "slides":["part-04.pdf",7],
 70 |         "slides_video":"https://www.bilibili.com/video/BV1JU4y1H7PC/",
 71 |         "notebook_video":"",
 72 |         "qa_video":""
 73 |     },
 74 |     {
 75 |         "title":"深层循环神经网络",
 76 |         "day_break":false,
 77 |         "book":"chapter_recurrent-modern/deep-rnn.html",
 78 |         "slides":["part-04.pdf",5],
 79 |         "slides_video":"https://www.bilibili.com/video/BV1JM4y1T7N4/",
 80 |         "notebook_video":"",
 81 |         "qa_video":""
 82 |     },
 83 |     {
 84 |         "title":"双向循环神经网络",
 85 |         "day_break":false,
 86 |         "book":"chapter_recurrent-modern/bi-rnn.html",
 87 |         "slides":["part-04.pdf",8],
 88 |         "slides_video":"https://www.bilibili.com/video/BV12X4y1c71W/",
 89 |         "notebook_video":"",
 90 |         "qa_video":""
 91 |     },
 92 |     {
 93 |         "title":"机器翻译与数据集",
 94 |         "day_break":true,
 95 |         "book":"chapter_recurrent-modern/machine-translation-and-dataset.html",
 96 |         "slides":["part-0.pdf",0],
 97 |         "slides_video":"https://www.bilibili.com/video/BV1H64y1s7TH/",
 98 |         "notebook_video":"",
 99 |         "qa_video":""
100 |     },
101 |     {
102 |         "title":"编码器-解码器结构",
103 |         "day_break":false,
104 |         "book":"chapter_recurrent-modern/encoder-decoder.html",
105 |         "slides":["part-04.pdf",5],
106 |         "slides_video":"https://www.bilibili.com/video/BV1c54y1E7YP/",
107 |         "notebook_video":"",
108 |         "qa_video":""
109 |     },
110 |     {
111 |         "title":"序列到序列学习（seq2seq）",
112 |         "day_break":false,
113 |         "book":"chapter_recurrent-modern/seq2seq.html",
114 |         "slides":["part-04.pdf",7],
115 |         "slides_video":"https://www.bilibili.com/video/BV16g411L7FG/",
116 |         "notebook_video":"",
117 |         "qa_video":""
118 |     },
119 |     {
120 |         "title":"束搜索",
121 |         "day_break":false,
122 |         "book":"chapter_recurrent-modern/beam-search.html",
123 |         "slides":["part-04.pdf",6],
124 |         "slides_video":"https://www.bilibili.com/video/BV1B44y1C7m1/",
125 |         "notebook_video":"",
126 |         "qa_video":""
127 |     }
128 | ]


--------------------------------------------------------------------------------
/_modules/part_3.md:
--------------------------------------------------------------------------------
  1 | ---
  2 | title: 循环神经网络
  3 | ---
  4 | 
  5 | 7月17日
  6 | 
  7 | : 序列模型
  8 |   : [<span class="iconfont icon-xiaoshuo-copy"></span>](https://zh-v2.d2l.ai/chapter_recurrent-neural-networks/sequence.html)
  9 |   : [<span class="iconfont icon-KeynoteOutline"></span>](assets/pdfs/part-3_1.pdf)
 10 |   : [<span class="iconfont icon-jupyter"></span>](assets/notebooks/chapter_recurrent-neural-networks/sequence.slides.html)
 11 |   : [<span style="font-size:130%"  class="iconfont icon-bilibili-fill"></span>](https://www.bilibili.com/video/BV1L44y1m768/)
 12 | 
 13 | : 文本预处理
 14 |   : [<span class="iconfont icon-xiaoshuo-copy"></span>](https://zh-v2.d2l.ai/chapter_recurrent-neural-networks/text-preprocessing.html)
 15 |   : &nbsp; 
 16 |   : [<span class="iconfont icon-jupyter"></span>](assets/notebooks/chapter_recurrent-neural-networks/text-preprocessing.slides.html)
 17 |   : [<span style="font-size:130%"  class="iconfont icon-bilibili-fill"></span>](https://www.bilibili.com/video/BV1Fo4y1Q79L/)
 18 | 
 19 | 
 20 | 7月18日
 21 | 
 22 | : 语言模型和数据集
 23 |   : [<span class="iconfont icon-xiaoshuo-copy"></span>](https://zh-v2.d2l.ai/chapter_recurrent-neural-networks/language-models-and-dataset.html)
 24 |   : [<span class="iconfont icon-KeynoteOutline"></span>](assets/pdfs/part-3_2.pdf)
 25 |   : [<span class="iconfont icon-jupyter"></span>](assets/notebooks/chapter_recurrent-neural-networks/language-models-and-dataset.slides.html)
 26 |   : [<span style="font-size:130%"  class="iconfont icon-bilibili-fill"></span>](https://www.bilibili.com/video/BV1ZX4y1F7K3/)
 27 | 
 28 | : 循环神经网络
 29 |   : [<span class="iconfont icon-xiaoshuo-copy"></span>](https://zh-v2.d2l.ai/chapter_recurrent-neural-networks/rnn.html)
 30 |   : [<span class="iconfont icon-KeynoteOutline"></span>](assets/pdfs/part-3_3.pdf)
 31 |   :  &nbsp; 
 32 |   : [<span style="font-size:130%"  class="iconfont icon-bilibili-fill"></span>](https://www.bilibili.com/video/BV1D64y1z7CA/)
 33 | 
 34 | 
 35 | 7月24日
 36 | 
 37 | : 循环神经网络的从零开始实现
 38 |   : [<span class="iconfont icon-xiaoshuo-copy"></span>](https://zh-v2.d2l.ai/chapter_recurrent-neural-networks/rnn-scratch.html)
 39 |   : &nbsp; 
 40 |   : [<span class="iconfont icon-jupyter"></span>](assets/notebooks/chapter_recurrent-neural-networks/rnn-scratch.slides.html)
 41 |   : [<span style="font-size:130%"  class="iconfont icon-bilibili-fill"></span>](https://www.bilibili.com/video/BV1kq4y1H7sw/)
 42 | 
 43 | : 循环神经网络的简洁实现
 44 |   : [<span class="iconfont icon-xiaoshuo-copy"></span>](https://zh-v2.d2l.ai/chapter_recurrent-neural-networks/rnn-concise.html)
 45 |   : &nbsp; 
 46 |   : [<span class="iconfont icon-jupyter"></span>](assets/notebooks/chapter_recurrent-neural-networks/rnn-concise.slides.html)
 47 |   : [<span style="font-size:130%"  class="iconfont icon-bilibili-fill"></span>](https://www.bilibili.com/video/BV1kq4y1H7sw?p=2)
 48 | 
 49 | 
 50 | 7月25日
 51 | 
 52 | : 门控循环单元（GRU）
 53 |   : [<span class="iconfont icon-xiaoshuo-copy"></span>](https://zh-v2.d2l.ai/chapter_recurrent-modern/gru.html)
 54 |   : [<span class="iconfont icon-KeynoteOutline"></span>](assets/pdfs/part-3_4.pdf)
 55 |   : [<span class="iconfont icon-jupyter"></span>](assets/notebooks/chapter_recurrent-modern/gru.slides.html)
 56 |   : [<span style="font-size:130%"  class="iconfont icon-bilibili-fill"></span>](https://www.bilibili.com/video/BV1mf4y157N2/)
 57 | 
 58 | : 长短期记忆网络（LSTM）
 59 |   : [<span class="iconfont icon-xiaoshuo-copy"></span>](https://zh-v2.d2l.ai/chapter_recurrent-modern/lstm.html)
 60 |   : [<span class="iconfont icon-KeynoteOutline"></span>](assets/pdfs/part-3_5.pdf)
 61 |   : [<span class="iconfont icon-jupyter"></span>](assets/notebooks/chapter_recurrent-modern/lstm.slides.html)
 62 |   : [<span style="font-size:130%"  class="iconfont icon-bilibili-fill"></span>](https://www.bilibili.com/video/BV1JU4y1H7PC/)
 63 | 
 64 | : 深层循环神经网络
 65 |   : [<span class="iconfont icon-xiaoshuo-copy"></span>](https://zh-v2.d2l.ai/chapter_recurrent-modern/deep-rnn.html)
 66 |   : [<span class="iconfont icon-KeynoteOutline"></span>](assets/pdfs/part-3_6.pdf)
 67 |   : [<span class="iconfont icon-jupyter"></span>](assets/notebooks/chapter_recurrent-modern/deep-rnn.slides.html)
 68 |   : [<span style="font-size:130%"  class="iconfont icon-bilibili-fill"></span>](https://www.bilibili.com/video/BV1JM4y1T7N4/)
 69 | 
 70 | : 双向循环神经网络
 71 |   : [<span class="iconfont icon-xiaoshuo-copy"></span>](https://zh-v2.d2l.ai/chapter_recurrent-modern/bi-rnn.html)
 72 |   : [<span class="iconfont icon-KeynoteOutline"></span>](assets/pdfs/part-3_7.pdf)
 73 |   : [<span class="iconfont icon-jupyter"></span>](assets/notebooks/chapter_recurrent-modern/bi-rnn.slides.html)
 74 |   : [<span style="font-size:130%"  class="iconfont icon-bilibili-fill"></span>](https://www.bilibili.com/video/BV12X4y1c71W/)
 75 | 
 76 | 
 77 | 7月31日
 78 | 
 79 | : **休课**{: .label .label-green }
 80 | 
 81 | 8月1日
 82 | 
 83 | : **休课**{: .label .label-green }
 84 | 
 85 | 8月7日
 86 | 
 87 | : 机器翻译与数据集
 88 |   : [<span class="iconfont icon-xiaoshuo-copy"></span>](https://zh-v2.d2l.ai/chapter_recurrent-modern/machine-translation-and-dataset.html)
 89 |   : &nbsp; 
 90 |   : [<span class="iconfont icon-jupyter"></span>](assets/notebooks/chapter_recurrent-modern/machine-translation-and-dataset.slides.html)
 91 |   : [<span style="font-size:130%"  class="iconfont icon-bilibili-fill"></span>](https://www.bilibili.com/video/BV1H64y1s7TH/)
 92 | 
 93 | : 编码器-解码器结构
 94 |   : [<span class="iconfont icon-xiaoshuo-copy"></span>](https://zh-v2.d2l.ai/chapter_recurrent-modern/encoder-decoder.html)
 95 |   : [<span class="iconfont icon-KeynoteOutline"></span>](assets/pdfs/part-3_8.pdf)
 96 |   : [<span class="iconfont icon-jupyter"></span>](assets/notebooks/chapter_recurrent-modern/encoder-decoder.slides.html)
 97 |   : [<span style="font-size:130%"  class="iconfont icon-bilibili-fill"></span>](https://www.bilibili.com/video/BV1c54y1E7YP/)
 98 | 
 99 | : 序列到序列学习（seq2seq）
100 |   : [<span class="iconfont icon-xiaoshuo-copy"></span>](https://zh-v2.d2l.ai/chapter_recurrent-modern/seq2seq.html)
101 |   : [<span class="iconfont icon-KeynoteOutline"></span>](assets/pdfs/part-3_9.pdf)
102 |   : [<span class="iconfont icon-jupyter"></span>](assets/notebooks/chapter_recurrent-modern/seq2seq.slides.html)
103 |   : [<span style="font-size:130%"  class="iconfont icon-bilibili-fill"></span>](https://www.bilibili.com/video/BV16g411L7FG/)
104 | 
105 | : 束搜索
106 |   : [<span class="iconfont icon-xiaoshuo-copy"></span>](https://zh-v2.d2l.ai/chapter_recurrent-modern/beam-search.html)
107 |   : [<span class="iconfont icon-KeynoteOutline"></span>](assets/pdfs/part-3_10.pdf)
108 |   :  &nbsp; 
109 |   : [<span style="font-size:130%"  class="iconfont icon-bilibili-fill"></span>](https://www.bilibili.com/video/BV1B44y1C7m1/)
110 | 
111 | 


--------------------------------------------------------------------------------
/_modules/part_4.json:
--------------------------------------------------------------------------------
  1 | [
  2 |     {
  3 |         "title":"注意力机制",
  4 |         "day_break":true,
  5 |         "book":"chapter_attention-mechanisms/nadaraya-waston.html",
  6 |         "slides":["part-05.pdf",7],
  7 |         "slides_video":"https://www.bilibili.com/video/BV1264y1i7R1/",
  8 |         "notebook_video":"",
  9 |         "qa_video":""
 10 |     },
 11 |     {
 12 |         "title":"注意力分数",
 13 |         "day_break":false,
 14 |         "book":"chapter_attention-mechanisms/attention-scoring-functions.html",
 15 |         "slides":["part-05.pdf",6],
 16 |         "slides_video":"https://www.bilibili.com/video/BV1Tb4y167rb/",
 17 |         "notebook_video":"",
 18 |         "qa_video":""
 19 |     },
 20 |     {
 21 |         "title":"使用注意力机制的seq2seq",
 22 |         "day_break":false,
 23 |         "book":"chapter_attention-mechanisms/bahdanau-attention.html",
 24 |         "slides":["part-05.pdf",4],
 25 |         "slides_video":"https://www.bilibili.com/video/BV1v44y1C7Tg/",
 26 |         "notebook_video":"",
 27 |         "qa_video":""
 28 |     },
 29 |     {
 30 |         "title":"自注意力和位置编码",
 31 |         "day_break":true,
 32 |         "book":"chapter_attention-mechanisms/self-attention-and-positional-encoding.html",
 33 |         "slides":["part-05.pdf",8],
 34 |         "slides_video":"https://www.bilibili.com/video/BV19o4y1m7mo/",
 35 |         "notebook_video":"",
 36 |         "qa_video":""
 37 |     },
 38 |     {
 39 |         "title":"Transformer",
 40 |         "day_break":false,
 41 |         "book":"chapter_attention-mechanisms/transformer.html",
 42 |         "slides":["part-05.pdf",10],
 43 |         "slides_video":"https://www.bilibili.com/video/BV1Kq4y1H7FL/",
 44 |         "notebook_video":"",
 45 |         "qa_video":""
 46 |     },
 47 |     {
 48 |         "title":"BERT",
 49 |         "day_break":true,
 50 |         "book":"chapter_natural-language-processing-pretraining/bert.html",
 51 |         "slides":["part-05.pdf",8],
 52 |         "slides_video":"https://www.bilibili.com/video/BV1yU4y1E7Ns/",
 53 |         "notebook_video":"",
 54 |         "qa_video":""
 55 |     },
 56 |     {
 57 |         "title":"BERT预训练数据集",
 58 |         "day_break":false,
 59 |         "book":"chapter_natural-language-processing-pretraining/bert-dataset.html",
 60 |         "slides":["part-0.pdf",0],
 61 |         "slides_video":"https://www.bilibili.com/video/BV1yU4y1E7Ns?p=2",
 62 |         "notebook_video":"",
 63 |         "qa_video":""
 64 |     },
 65 |     {
 66 |         "title":"预训练BERT",
 67 |         "day_break":false,
 68 |         "book":"chapter_natural-language-processing-pretraining/bert-pretraining.html",
 69 |         "slides":["part-0.pdf",0],
 70 |         "slides_video":"https://www.bilibili.com/video/BV1yU4y1E7Ns?p=3",
 71 |         "notebook_video":"",
 72 |         "qa_video":""
 73 |     },
 74 |     {
 75 |         "title":"微调BERT",
 76 |         "day_break":false,
 77 |         "book":"chapter_natural-language-processing-applications/finetuning-bert.html",
 78 |         "slides":["part-05.pdf",6],
 79 |         "slides_video":"https://www.bilibili.com/video/BV15L4y1v7ts",
 80 |         "notebook_video":"",
 81 |         "qa_video":""
 82 |     },
 83 |     {
 84 |         "title":"自然语言推理和数据集",
 85 |         "day_break":false,
 86 |         "book":"chapter_natural-language-processing-applications/natural-language-inference-and-dataset.html",
 87 |         "slides":["part-0.pdf",0],
 88 |         "slides_video":"https://www.bilibili.com/video/BV15L4y1v7ts?p=2",
 89 |         "notebook_video":"",
 90 |         "qa_video":""
 91 |     },
 92 |     {
 93 |         "title":"自然语言推理：微调BERT",
 94 |         "day_break":false,
 95 |         "book":"chapter_natural-language-processing-applications/natural-language-inference-bert.html",
 96 |         "slides":["part-0.pdf",0],
 97 |         "slides_video":"https://www.bilibili.com/video/BV15L4y1v7ts?p=3",
 98 |         "notebook_video":"",
 99 |         "qa_video":""
100 |     },
101 |     {
102 |         "title":"**竞赛**{: .label  } 目标检测总结",
103 |         "day_break":false,
104 |         "book":"",
105 |         "slides":["part-07.pdf",6],
106 |         "slides_video":"https://www.bilibili.com/video/BV13b4y1m7y8",
107 |         "notebook_video":"",
108 |         "qa_video":""
109 |     },
110 |     {
111 |         "title":"优化算法",
112 |         "day_break":true,
113 |         "book":"chapter_optimization/index.html",
114 |         "slides":["part-07.pdf",14],
115 |         "slides_video":"https://www.bilibili.com/video/BV1bP4y1p7Gq",
116 |         "notebook_video":"",
117 |         "qa_video":""
118 |     },
119 |     {
120 |         "title":"课程总结和进阶学习",
121 |         "day_break":false,
122 |         "book":"",
123 |         "slides":["part-07.pdf",11],
124 |         "slides_video":"https://www.bilibili.com/video/BV1AL4y1Y7gu",
125 |         "notebook_video":"",
126 |         "qa_video":""
127 |     }
128 | ]


--------------------------------------------------------------------------------
/_modules/part_4.md:
--------------------------------------------------------------------------------
  1 | ---
  2 | title: 注意力机制
  3 | ---
  4 | 
  5 | 8月8日
  6 | 
  7 | : 注意力机制
  8 |   : [<span class="iconfont icon-xiaoshuo-copy"></span>](https://zh-v2.d2l.ai/chapter_attention-mechanisms/nadaraya-waston.html)
  9 |   : [<span class="iconfont icon-KeynoteOutline"></span>](assets/pdfs/part-4_1.pdf)
 10 |   : [<span class="iconfont icon-jupyter"></span>](assets/notebooks/chapter_attention-mechanisms/nadaraya-waston.slides.html)
 11 |   : [<span style="font-size:130%"  class="iconfont icon-bilibili-fill"></span>](https://www.bilibili.com/video/BV1264y1i7R1/)
 12 | 
 13 | : 注意力分数
 14 |   : [<span class="iconfont icon-xiaoshuo-copy"></span>](https://zh-v2.d2l.ai/chapter_attention-mechanisms/attention-scoring-functions.html)
 15 |   : [<span class="iconfont icon-KeynoteOutline"></span>](assets/pdfs/part-4_2.pdf)
 16 |   : [<span class="iconfont icon-jupyter"></span>](assets/notebooks/chapter_attention-mechanisms/attention-scoring-functions.slides.html)
 17 |   : [<span style="font-size:130%"  class="iconfont icon-bilibili-fill"></span>](https://www.bilibili.com/video/BV1Tb4y167rb/)
 18 | 
 19 | : 使用注意力机制的seq2seq
 20 |   : [<span class="iconfont icon-xiaoshuo-copy"></span>](https://zh-v2.d2l.ai/chapter_attention-mechanisms/bahdanau-attention.html)
 21 |   : [<span class="iconfont icon-KeynoteOutline"></span>](assets/pdfs/part-4_3.pdf)
 22 |   : [<span class="iconfont icon-jupyter"></span>](assets/notebooks/chapter_attention-mechanisms/bahdanau-attention.slides.html)
 23 |   : [<span style="font-size:130%"  class="iconfont icon-bilibili-fill"></span>](https://www.bilibili.com/video/BV1v44y1C7Tg/)
 24 | 
 25 | 
 26 | 8月14日
 27 | 
 28 | : 自注意力和位置编码
 29 |   : [<span class="iconfont icon-xiaoshuo-copy"></span>](https://zh-v2.d2l.ai/chapter_attention-mechanisms/self-attention-and-positional-encoding.html)
 30 |   : [<span class="iconfont icon-KeynoteOutline"></span>](assets/pdfs/part-4_4.pdf)
 31 |   : [<span class="iconfont icon-jupyter"></span>](assets/notebooks/chapter_attention-mechanisms/self-attention-and-positional-encoding.slides.html)
 32 |   : [<span style="font-size:130%"  class="iconfont icon-bilibili-fill"></span>](https://www.bilibili.com/video/BV19o4y1m7mo/)
 33 | 
 34 | : Transformer
 35 |   : [<span class="iconfont icon-xiaoshuo-copy"></span>](https://zh-v2.d2l.ai/chapter_attention-mechanisms/transformer.html)
 36 |   : [<span class="iconfont icon-KeynoteOutline"></span>](assets/pdfs/part-4_5.pdf)
 37 |   : [<span class="iconfont icon-jupyter"></span>](assets/notebooks/chapter_attention-mechanisms/transformer.slides.html)
 38 |   : [<span style="font-size:130%"  class="iconfont icon-bilibili-fill"></span>](https://www.bilibili.com/video/BV1Kq4y1H7FL/)
 39 | 
 40 | 
 41 | 8月15日
 42 | 
 43 | : BERT
 44 |   : [<span class="iconfont icon-xiaoshuo-copy"></span>](https://zh-v2.d2l.ai/chapter_natural-language-processing-pretraining/bert.html)
 45 |   : [<span class="iconfont icon-KeynoteOutline"></span>](assets/pdfs/part-4_6.pdf)
 46 |   : [<span class="iconfont icon-jupyter"></span>](assets/notebooks/chapter_natural-language-processing-pretraining/bert.slides.html)
 47 |   : [<span style="font-size:130%"  class="iconfont icon-bilibili-fill"></span>](https://www.bilibili.com/video/BV1yU4y1E7Ns/)
 48 | 
 49 | : BERT预训练数据集
 50 |   : [<span class="iconfont icon-xiaoshuo-copy"></span>](https://zh-v2.d2l.ai/chapter_natural-language-processing-pretraining/bert-dataset.html)
 51 |   : &nbsp; 
 52 |   : [<span class="iconfont icon-jupyter"></span>](assets/notebooks/chapter_natural-language-processing-pretraining/bert-dataset.slides.html)
 53 |   : [<span style="font-size:130%"  class="iconfont icon-bilibili-fill"></span>](https://www.bilibili.com/video/BV1yU4y1E7Ns?p=2)
 54 | 
 55 | : 预训练BERT
 56 |   : [<span class="iconfont icon-xiaoshuo-copy"></span>](https://zh-v2.d2l.ai/chapter_natural-language-processing-pretraining/bert-pretraining.html)
 57 |   : &nbsp; 
 58 |   : [<span class="iconfont icon-jupyter"></span>](assets/notebooks/chapter_natural-language-processing-pretraining/bert-pretraining.slides.html)
 59 |   : [<span style="font-size:130%"  class="iconfont icon-bilibili-fill"></span>](https://www.bilibili.com/video/BV1yU4y1E7Ns?p=3)
 60 | 
 61 | : 微调BERT
 62 |   : [<span class="iconfont icon-xiaoshuo-copy"></span>](https://zh-v2.d2l.ai/chapter_natural-language-processing-applications/finetuning-bert.html)
 63 |   : [<span class="iconfont icon-KeynoteOutline"></span>](assets/pdfs/part-4_7.pdf)
 64 |   :  &nbsp; 
 65 |   : [<span style="font-size:130%"  class="iconfont icon-bilibili-fill"></span>](https://www.bilibili.com/video/BV15L4y1v7ts)
 66 | 
 67 | : 自然语言推理和数据集
 68 |   : [<span class="iconfont icon-xiaoshuo-copy"></span>](https://zh-v2.d2l.ai/chapter_natural-language-processing-applications/natural-language-inference-and-dataset.html)
 69 |   : &nbsp; 
 70 |   : [<span class="iconfont icon-jupyter"></span>](assets/notebooks/chapter_natural-language-processing-applications/natural-language-inference-and-dataset.slides.html)
 71 |   : [<span style="font-size:130%"  class="iconfont icon-bilibili-fill"></span>](https://www.bilibili.com/video/BV15L4y1v7ts?p=2)
 72 | 
 73 | : 自然语言推理：微调BERT
 74 |   : [<span class="iconfont icon-xiaoshuo-copy"></span>](https://zh-v2.d2l.ai/chapter_natural-language-processing-applications/natural-language-inference-bert.html)
 75 |   : &nbsp; 
 76 |   : [<span class="iconfont icon-jupyter"></span>](assets/notebooks/chapter_natural-language-processing-applications/natural-language-inference-bert.slides.html)
 77 |   : [<span style="font-size:130%"  class="iconfont icon-bilibili-fill"></span>](https://www.bilibili.com/video/BV15L4y1v7ts?p=3)
 78 | 
 79 | : **竞赛**{: .label  } 目标检测总结
 80 |   : &nbsp; 
 81 |   : [<span class="iconfont icon-KeynoteOutline"></span>](assets/pdfs/part-4_1.pdf)
 82 |   :  &nbsp; 
 83 |   : [<span style="font-size:130%"  class="iconfont icon-bilibili-fill"></span>](https://www.bilibili.com/video/BV13b4y1m7y8)
 84 | 
 85 | 
 86 | 8月21日
 87 | 
 88 | : 优化算法
 89 |   : [<span class="iconfont icon-xiaoshuo-copy"></span>](https://zh-v2.d2l.ai/chapter_optimization/index.html)
 90 |   : [<span class="iconfont icon-KeynoteOutline"></span>](assets/pdfs/part-4_2.pdf)
 91 |   :  &nbsp; 
 92 |   : [<span style="font-size:130%"  class="iconfont icon-bilibili-fill"></span>](https://www.bilibili.com/video/BV1bP4y1p7Gq)
 93 | 
 94 | : 课程总结和进阶学习
 95 |   : &nbsp; 
 96 |   : [<span class="iconfont icon-KeynoteOutline"></span>](assets/pdfs/part-4_3.pdf)
 97 |   :  &nbsp; 
 98 |   : [<span style="font-size:130%"  class="iconfont icon-bilibili-fill"></span>](https://www.bilibili.com/video/BV1AL4y1Y7gu)
 99 | 
100 | 


--------------------------------------------------------------------------------
/_sass/color_schemes/d2l.scss:
--------------------------------------------------------------------------------
1 | $link-color: rgb(33,150,243);
2 | 


--------------------------------------------------------------------------------
/_sass/custom/announcement.scss:
--------------------------------------------------------------------------------
 1 | .announcement {
 2 |   @extend %card;
 3 | 
 4 |   h1, h2 {
 5 |     @extend .text-gamma;
 6 |   }
 7 | 
 8 |   .announcement-meta {
 9 |     @extend .text-epsilon;
10 |   }
11 | }
12 | 


--------------------------------------------------------------------------------
/_sass/custom/card.scss:
--------------------------------------------------------------------------------
 1 | @mixin abstract-card() {
 2 |   box-shadow: 0 1px 3px rgba(0, 0, 0, 0.07), 0 4px 14px rgba(0, 0, 0, 0.05);
 3 |   margin: $sp-4 (-$gutter-spacing-sm);
 4 | 
 5 |   @include mq(md) {
 6 |     border-radius: $border-radius;
 7 |     margin: $sp-4 0;
 8 |   }
 9 | }
10 | 
11 | %card {
12 |   @include abstract-card();
13 |   display: flex;
14 |   flex-direction: column;
15 |   min-width: 0;
16 |   padding: 0 $sp-4;
17 |   position: relative;
18 |   word-wrap: break-word;
19 | 
20 |   >:first-child {
21 |     border-top: none !important;
22 |   }
23 | 
24 |   >:last-child {
25 |     border-bottom: none !important;
26 |   }
27 | 
28 |   .label {
29 |     border-radius: $border-radius;
30 |     margin-left: 0;
31 |     user-select: none;
32 |   }
33 | }
34 | 


--------------------------------------------------------------------------------
/_sass/custom/custom.scss:
--------------------------------------------------------------------------------
  1 | // Just the Class dependencies
  2 | @import 'card';
  3 | 
  4 | // Just the Class styles
  5 | @import 'announcement';
  6 | @import 'module';
  7 | @import 'schedule';
  8 | @import 'staffer';
  9 | 
 10 | // Overrides
 11 | a abbr[title] {
 12 |   border-bottom: none;
 13 | }
 14 | 
 15 | abbr[title] {
 16 |   text-decoration: none;
 17 | }
 18 | 
 19 | code {
 20 |   font-size: 14px;
 21 |   padding: 0.2em 0.4em;
 22 |   border: none;
 23 | }
 24 | 
 25 | div.highlighter-rouge[overlay] {
 26 |   position: relative;
 27 | 
 28 |   &::after {
 29 |     @extend .label, .text-grey-dk-100;
 30 | 
 31 |     background-color: $white;
 32 |     border-radius: $border-radius;
 33 |     bottom: $sp-2;
 34 |     content: attr(overlay);
 35 |     position: absolute;
 36 |     right: 0;
 37 |     user-select: none;
 38 |   }
 39 | }
 40 | 
 41 | details {
 42 |   margin: 0 40px 1em;
 43 | }
 44 | 
 45 | h1, h2, h3, h4, h5, h6 {
 46 |   align-items: center;
 47 |   display: flex;
 48 | }
 49 | 
 50 | iframe,
 51 | summary {
 52 |   max-width: 100%;
 53 | }
 54 | 
 55 | summary {
 56 |   @extend .btn, .btn-outline;
 57 | }
 58 | 
 59 | .main-content-wrap {
 60 |   max-width: $content-width;
 61 |   margin: auto;
 62 | }
 63 | 
 64 | .main-content {
 65 |   a {
 66 |     overflow-wrap: anywhere;
 67 |     white-space: normal;
 68 |   }
 69 | 
 70 |   dl {
 71 |     display: block;
 72 |     grid-template-columns: none;
 73 |   }
 74 | 
 75 |   dt {
 76 |     font-weight: normal;
 77 |     text-align: start;
 78 | 
 79 |     &::after {
 80 |       content: normal;
 81 |     }
 82 |   }
 83 | 
 84 |   dd {
 85 |     font-weight: normal;
 86 | 
 87 |     + dt {
 88 |       margin-top: 1em;
 89 |     }
 90 |   }
 91 | 
 92 |   .katex {
 93 |     font-size: 1.1em;
 94 |   }
 95 | }
 96 | 
 97 | [style*="--aspect-ratio"] > :first-child {
 98 |   width: 100%;
 99 | }
100 | 
101 | [style*="--aspect-ratio"] > img {
102 |   height: auto;
103 | }
104 | 
105 | @supports (--custom:property) {
106 |   [style*="--aspect-ratio"] {
107 |     position: relative;
108 |   }
109 | 
110 |   [style*="--aspect-ratio"]::before {
111 |     content: "";
112 |     display: block;
113 |     padding-bottom: calc(100% / (var(--aspect-ratio)));
114 |   }
115 | 
116 |   [style*="--add-height"]::before {
117 |     padding-bottom: calc(100% / (var(--aspect-ratio)) + (var(--add-height)));
118 |   }
119 | 
120 |   [style*="--aspect-ratio"] > :first-child {
121 |     position: absolute;
122 |     top: 0;
123 |     left: 0;
124 |     height: 100%;
125 |   }
126 | }
127 | 
128 | .responsive-video-container {
129 |   position: relative;
130 |   margin-bottom: 1em;
131 |   padding-bottom: 56.25%;
132 |   height: 0;
133 |   overflow: hidden;
134 |   max-width: 100%;
135 | 
136 |   iframe,
137 |   object,
138 |   embed {
139 |     position: absolute;
140 |     top: 0;
141 |     left: 0;
142 |     width: 100%;
143 |     height: 100%;
144 |   }
145 | }


--------------------------------------------------------------------------------
/_sass/custom/module.scss:
--------------------------------------------------------------------------------
  1 | .main-content .module,
  2 | .module {
  3 |   @extend %card;
  4 | 
  5 |   h1,
  6 |   h2,
  7 |   h3,
  8 |   h4,
  9 |   h5,
 10 |   h6 {
 11 |     &:first-child {
 12 |       margin-top: $sp-4;
 13 |     }
 14 |   }
 15 | 
 16 |   >dl {
 17 |     border-bottom: $border $border-color;
 18 |     border-top: $border $border-color;
 19 |     display: grid;
 20 |     grid-template-columns: max-content 1fr;
 21 |     margin: $sp-2 (-$sp-4);
 22 | 
 23 |     &:first-child {
 24 |       margin-top: 0;
 25 |     }
 26 | 
 27 |     &:last-child {
 28 |       margin-bottom: 0;
 29 |     }
 30 | 
 31 |     @include mq(lg) {
 32 |       grid-template-columns: 1fr 7fr;
 33 |     }
 34 | 
 35 |     %module-item {
 36 |       margin: 0;
 37 |       padding: $sp-2;
 38 | 
 39 |       @include mq(sm) {
 40 |         padding: $sp-2 $sp-4;
 41 |       }
 42 |     }
 43 | 
 44 |     >dt {
 45 |       @extend %module-item;
 46 |       border-top: $border $border-color;
 47 |       font-weight: normal;
 48 |       text-align: right;
 49 |       width: 100px;
 50 | 
 51 |       +dd {
 52 |         border-top: $border $border-color;
 53 |       }
 54 | 
 55 |       &:first-child {
 56 |         border-top: none;
 57 | 
 58 |         +dd {
 59 |           border-top: none;
 60 |         }
 61 |       }
 62 | 
 63 |       &::after {
 64 |         content: ":";
 65 |       }
 66 |     }
 67 | 
 68 |     >dd {
 69 |       @extend %module-item;
 70 | 
 71 | 
 72 |       a {
 73 |         background-image: none;
 74 |       }
 75 |       i {
 76 |         padding-right: 0.2em;
 77 |         color: gray;
 78 |         &:hover {
 79 |           color: #2196f3;
 80 |         }
 81 |       }
 82 |       +dd {
 83 |         padding-top: 0;
 84 |       }
 85 | 
 86 |       ol, ul, dl {
 87 |         margin: 0;
 88 |       }
 89 |       @media only screen and (max-width: 500px) {
 90 |         dl {
 91 |           display: flex;
 92 |           flex-direction: row;
 93 |           flex-flow: row wrap;
 94 | 
 95 |           dt {
 96 |             width: 100%;
 97 |             margin: 0;
 98 |             padding-bottom: 0.2em;
 99 |           }
100 | 
101 |           dd {
102 |             font-size: 80%;
103 |             width: 10%;
104 |             margin: 0;
105 |             padding-bottom: 0.4em;
106 |           }
107 |         }
108 | 
109 |       }
110 |       @media only screen and (min-width: 500px) {
111 |         dl {
112 |           display: flex;
113 |           flex-direction: column;
114 | 
115 |           @include mq(sm) {
116 |             flex-direction: row;
117 |           }
118 | 
119 |           dt {
120 |             flex: 0 0 66%;
121 |             margin: 0;
122 |           }
123 | 
124 |           dd {
125 |             font-size: 80%;
126 |             flex: 0 0 8%;
127 |             margin: 0;
128 |           }
129 |         }
130 |       }
131 | 
132 |     }
133 |   }
134 | }
135 | 


--------------------------------------------------------------------------------
/_sass/custom/schedule.scss:
--------------------------------------------------------------------------------
  1 | .schedule {
  2 |   @include abstract-card();
  3 |   overflow-x: scroll;
  4 |   position: relative;
  5 | 
  6 |   li::before {
  7 |     display: none;
  8 |   }
  9 | 
 10 |   ul.schedule-timeline,
 11 |   ul.schedule-group,
 12 |   ul.schedule-events {
 13 |     margin-top: 0;
 14 |     padding-left: 0;
 15 |   }
 16 | 
 17 |   ul.schedule-timeline {
 18 |     margin: 40px auto 0;
 19 |     position: absolute;
 20 |     width: 100%;
 21 |   }
 22 | 
 23 |   .schedule-time {
 24 |     @extend .fs-2;
 25 |     color: $grey-dk-000;
 26 |     height: 40px;
 27 |     margin: 0;
 28 |     padding: $sp-2;
 29 |     position: relative;
 30 | 
 31 |     &::after {
 32 |       background-color: $border-color;
 33 |       content: '';
 34 |       height: 1px;
 35 |       left: 0;
 36 |       position: absolute;
 37 |       top: 0;
 38 |       width: 100%;
 39 |     }
 40 |   }
 41 | 
 42 |   .schedule-group {
 43 |     display: flex;
 44 |     margin-bottom: 0;
 45 |     position: relative;
 46 |   }
 47 | 
 48 |   .schedule-day {
 49 |     border-left: $border $border-color;
 50 |     flex: 1 0 0;
 51 |     margin: 0;
 52 |     min-width: 130px;
 53 | 
 54 |     &:first-of-type {
 55 |       border-left: 0;
 56 |     }
 57 |   }
 58 | 
 59 |   h2.schedule-header {
 60 |     align-items: center;
 61 |     display: flex;
 62 |     font-size: 18px !important;
 63 |     height: 40px;
 64 |     justify-content: center;
 65 |     margin: 0;
 66 |   }
 67 | 
 68 |   .schedule-events {
 69 |     display: flex;
 70 |     padding: 0;
 71 |     position: relative;
 72 |   }
 73 | 
 74 |   .schedule-event {
 75 |     background-color: $grey-dk-000;
 76 |     border-radius: $border-radius;
 77 |     box-shadow: 0 10px 20px rgba(0, 0, 0, .1), inset 0 -3px 0 rgba(0, 0, 0, .2);
 78 |     color: $white;
 79 |     float: left;
 80 |     height: 100%;
 81 |     margin: 0;
 82 |     padding: $sp-1 $sp-2;
 83 |     position: absolute;
 84 |     width: 100%;
 85 | 
 86 |     .name {
 87 |       @extend .fs-3, .fw-700;
 88 |     }
 89 | 
 90 |     .time,
 91 |     .location {
 92 |       @extend .fs-2;
 93 |     }
 94 | 
 95 |     &.lecture {
 96 |       background-color: $grey-dk-000;
 97 |     }
 98 | 
 99 |     &.section {
100 |       background-color: $purple-000;
101 |     }
102 | 
103 |     &.office-hours {
104 |       background-color: $blue-000;
105 |     }
106 |   }
107 | }
108 | 


--------------------------------------------------------------------------------
/_sass/custom/staffer.scss:
--------------------------------------------------------------------------------
 1 | .staffer {
 2 |   display: block;
 3 |   margin: $sp-4;
 4 |   float: left;
 5 |   width: 100%;
 6 | 
 7 |   .staffer-info {
 8 |     display: block;
 9 |   }
10 | 
11 |   .staffer-image {
12 |     border-radius: 50%;
13 |     height: 100px;
14 |     margin-right: 1.5em;
15 |     float: left;
16 | 
17 |   }
18 | 
19 |   p,
20 |   .staffer-name {
21 |     margin: $sp-1 !important;
22 |   }
23 | 
24 |   .staffer-pronouns {
25 |     @extend .label, .text-grey-dk-100, .bg-grey-lt-200;
26 |   }
27 | 
28 |   .staffer-meta {
29 |     @extend .text-grey-dk-000;
30 |   }
31 | }
32 | 


--------------------------------------------------------------------------------
/_staffers/mu.md:
--------------------------------------------------------------------------------
1 | ---
2 | name: 李沐
3 | website: https://github.com/mli
4 | photo: mu.jpg
5 | ---
6 | 
7 | AWS 资深首席科学家<br>美国卡内基梅隆大学计算机系博士


--------------------------------------------------------------------------------
/announcements.md:
--------------------------------------------------------------------------------
 1 | ---
 2 | layout: page
 3 | title: 课程公告
 4 | nav_exclude: true
 5 | description: A feed containing all of the class announcements.
 6 | ---
 7 | 
 8 | # 课程公告
 9 | 
10 | {% assign announcements = site.announcements | reverse %}
11 | {% for announcement in announcements %}
12 | {{ announcement }}
13 | {% endfor %}
14 | 


--------------------------------------------------------------------------------
/assets/images/logo-with-text.png:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/d2l-ai/courses-zh-v2/3d1b8b89eae0c836a24420051d33f2876a32e4d4/assets/images/logo-with-text.png


--------------------------------------------------------------------------------
/assets/images/mu.jpg:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/d2l-ai/courses-zh-v2/3d1b8b89eae0c836a24420051d33f2876a32e4d4/assets/images/mu.jpg


--------------------------------------------------------------------------------
/assets/pdfs/part-0_1.pdf:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/d2l-ai/courses-zh-v2/3d1b8b89eae0c836a24420051d33f2876a32e4d4/assets/pdfs/part-0_1.pdf


--------------------------------------------------------------------------------
/assets/pdfs/part-0_2.pdf:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/d2l-ai/courses-zh-v2/3d1b8b89eae0c836a24420051d33f2876a32e4d4/assets/pdfs/part-0_2.pdf


--------------------------------------------------------------------------------
/assets/pdfs/part-0_3.pdf:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/d2l-ai/courses-zh-v2/3d1b8b89eae0c836a24420051d33f2876a32e4d4/assets/pdfs/part-0_3.pdf


--------------------------------------------------------------------------------
/assets/pdfs/part-0_4.pdf:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/d2l-ai/courses-zh-v2/3d1b8b89eae0c836a24420051d33f2876a32e4d4/assets/pdfs/part-0_4.pdf


--------------------------------------------------------------------------------
/assets/pdfs/part-0_5.pdf:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/d2l-ai/courses-zh-v2/3d1b8b89eae0c836a24420051d33f2876a32e4d4/assets/pdfs/part-0_5.pdf


--------------------------------------------------------------------------------
/assets/pdfs/part-0_6.pdf:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/d2l-ai/courses-zh-v2/3d1b8b89eae0c836a24420051d33f2876a32e4d4/assets/pdfs/part-0_6.pdf


--------------------------------------------------------------------------------
/assets/pdfs/part-0_7.pdf:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/d2l-ai/courses-zh-v2/3d1b8b89eae0c836a24420051d33f2876a32e4d4/assets/pdfs/part-0_7.pdf


--------------------------------------------------------------------------------
/assets/pdfs/part-0_8.pdf:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/d2l-ai/courses-zh-v2/3d1b8b89eae0c836a24420051d33f2876a32e4d4/assets/pdfs/part-0_8.pdf


--------------------------------------------------------------------------------
/contents.json:
--------------------------------------------------------------------------------
   1 | [
   2 |     {
   3 |         "title":"Preface",
   4 |         "day_break":false,
   5 |         "book":"chapter_preface/index.html",
   6 |         "slides":["part-0.pdf",0],
   7 |         "slides_video":"",
   8 |         "notebook_video":"",
   9 |         "qa_video":""
  10 |     },
  11 |     {
  12 |         "title":"Installation",
  13 |         "day_break":false,
  14 |         "book":"chapter_installation/index.html",
  15 |         "slides":["part-0.pdf",0],
  16 |         "slides_video":"",
  17 |         "notebook_video":"",
  18 |         "qa_video":""
  19 |     },
  20 |     {
  21 |         "title":"Notation",
  22 |         "day_break":false,
  23 |         "book":"chapter_notation/index.html",
  24 |         "slides":["part-0.pdf",0],
  25 |         "slides_video":"",
  26 |         "notebook_video":"",
  27 |         "qa_video":""
  28 |     },
  29 |     {
  30 |         "title":"Introduction",
  31 |         "day_break":false,
  32 |         "book":"chapter_introduction/index.html",
  33 |         "slides":["part-0.pdf",0],
  34 |         "slides_video":"",
  35 |         "notebook_video":"",
  36 |         "qa_video":""
  37 |     },
  38 |     {
  39 |         "title":"Preliminaries",
  40 |         "day_break":false,
  41 |         "book":"chapter_preliminaries/index.html",
  42 |         "slides":["part-0.pdf",0],
  43 |         "slides_video":"",
  44 |         "notebook_video":"",
  45 |         "qa_video":""
  46 |     },
  47 |     {
  48 |         "title":"Data Manipulation",
  49 |         "day_break":false,
  50 |         "book":"chapter_preliminaries/ndarray.html",
  51 |         "slides":["part-0.pdf",0],
  52 |         "slides_video":"",
  53 |         "notebook_video":"",
  54 |         "qa_video":""
  55 |     },
  56 |     {
  57 |         "title":"Data Preprocessing",
  58 |         "day_break":false,
  59 |         "book":"chapter_preliminaries/pandas.html",
  60 |         "slides":["part-0.pdf",0],
  61 |         "slides_video":"",
  62 |         "notebook_video":"",
  63 |         "qa_video":""
  64 |     },
  65 |     {
  66 |         "title":"Linear Algebra",
  67 |         "day_break":false,
  68 |         "book":"chapter_preliminaries/linear-algebra.html",
  69 |         "slides":["part-0.pdf",0],
  70 |         "slides_video":"",
  71 |         "notebook_video":"",
  72 |         "qa_video":""
  73 |     },
  74 |     {
  75 |         "title":"Calculus",
  76 |         "day_break":false,
  77 |         "book":"chapter_preliminaries/calculus.html",
  78 |         "slides":["part-0.pdf",0],
  79 |         "slides_video":"",
  80 |         "notebook_video":"",
  81 |         "qa_video":""
  82 |     },
  83 |     {
  84 |         "title":"Automatic Differentiation",
  85 |         "day_break":false,
  86 |         "book":"chapter_preliminaries/autograd.html",
  87 |         "slides":["part-0.pdf",0],
  88 |         "slides_video":"",
  89 |         "notebook_video":"",
  90 |         "qa_video":""
  91 |     },
  92 |     {
  93 |         "title":"Probability",
  94 |         "day_break":false,
  95 |         "book":"chapter_preliminaries/probability.html",
  96 |         "slides":["part-0.pdf",0],
  97 |         "slides_video":"",
  98 |         "notebook_video":"",
  99 |         "qa_video":""
 100 |     },
 101 |     {
 102 |         "title":"Documentation",
 103 |         "day_break":false,
 104 |         "book":"chapter_preliminaries/lookup-api.html",
 105 |         "slides":["part-0.pdf",0],
 106 |         "slides_video":"",
 107 |         "notebook_video":"",
 108 |         "qa_video":""
 109 |     },
 110 |     {
 111 |         "title":"Linear Neural Networks",
 112 |         "day_break":false,
 113 |         "book":"chapter_linear-networks/index.html",
 114 |         "slides":["part-0.pdf",0],
 115 |         "slides_video":"",
 116 |         "notebook_video":"",
 117 |         "qa_video":""
 118 |     },
 119 |     {
 120 |         "title":"Linear Regression",
 121 |         "day_break":false,
 122 |         "book":"chapter_linear-networks/linear-regression.html",
 123 |         "slides":["part-0.pdf",0],
 124 |         "slides_video":"",
 125 |         "notebook_video":"",
 126 |         "qa_video":""
 127 |     },
 128 |     {
 129 |         "title":"The D2L APIs",
 130 |         "day_break":false,
 131 |         "book":"chapter_linear-networks/api.html",
 132 |         "slides":["part-0.pdf",0],
 133 |         "slides_video":"",
 134 |         "notebook_video":"",
 135 |         "qa_video":""
 136 |     },
 137 |     {
 138 |         "title":"Synthetic Regression Data",
 139 |         "day_break":false,
 140 |         "book":"chapter_linear-networks/synthetic-regression-data.html",
 141 |         "slides":["part-0.pdf",0],
 142 |         "slides_video":"",
 143 |         "notebook_video":"",
 144 |         "qa_video":""
 145 |     },
 146 |     {
 147 |         "title":"Linear Regression Implementation from Scratch",
 148 |         "day_break":false,
 149 |         "book":"chapter_linear-networks/linear-regression-scratch.html",
 150 |         "slides":["part-0.pdf",0],
 151 |         "slides_video":"",
 152 |         "notebook_video":"",
 153 |         "qa_video":""
 154 |     },
 155 |     {
 156 |         "title":"Concise Implementation of Linear Regression",
 157 |         "day_break":false,
 158 |         "book":"chapter_linear-networks/linear-regression-concise.html",
 159 |         "slides":["part-0.pdf",0],
 160 |         "slides_video":"",
 161 |         "notebook_video":"",
 162 |         "qa_video":""
 163 |     },
 164 |     {
 165 |         "title":"Softmax Regression",
 166 |         "day_break":false,
 167 |         "book":"chapter_linear-networks/softmax-regression.html",
 168 |         "slides":["part-0.pdf",0],
 169 |         "slides_video":"",
 170 |         "notebook_video":"",
 171 |         "qa_video":""
 172 |     },
 173 |     {
 174 |         "title":"The Image Classification Dataset",
 175 |         "day_break":false,
 176 |         "book":"chapter_linear-networks/image-classification-dataset.html",
 177 |         "slides":["part-0.pdf",0],
 178 |         "slides_video":"",
 179 |         "notebook_video":"",
 180 |         "qa_video":""
 181 |     },
 182 |     {
 183 |         "title":"The Base Classification Model",
 184 |         "day_break":false,
 185 |         "book":"chapter_linear-networks/classification.html",
 186 |         "slides":["part-0.pdf",0],
 187 |         "slides_video":"",
 188 |         "notebook_video":"",
 189 |         "qa_video":""
 190 |     },
 191 |     {
 192 |         "title":"Softmax Regression Implementation from Scratch",
 193 |         "day_break":false,
 194 |         "book":"chapter_linear-networks/softmax-regression-scratch.html",
 195 |         "slides":["part-0.pdf",0],
 196 |         "slides_video":"",
 197 |         "notebook_video":"",
 198 |         "qa_video":""
 199 |     },
 200 |     {
 201 |         "title":"Concise Implementation of Softmax Regression",
 202 |         "day_break":false,
 203 |         "book":"chapter_linear-networks/softmax-regression-concise.html",
 204 |         "slides":["part-0.pdf",0],
 205 |         "slides_video":"",
 206 |         "notebook_video":"",
 207 |         "qa_video":""
 208 |     },
 209 |     {
 210 |         "title":"Multilayer Perceptrons",
 211 |         "day_break":false,
 212 |         "book":"chapter_multilayer-perceptrons/index.html",
 213 |         "slides":["part-0.pdf",0],
 214 |         "slides_video":"",
 215 |         "notebook_video":"",
 216 |         "qa_video":""
 217 |     },
 218 |     {
 219 |         "title":"Multilayer Perceptrons",
 220 |         "day_break":false,
 221 |         "book":"chapter_multilayer-perceptrons/mlp.html",
 222 |         "slides":["part-0.pdf",0],
 223 |         "slides_video":"",
 224 |         "notebook_video":"",
 225 |         "qa_video":""
 226 |     },
 227 |     {
 228 |         "title":"Multilayer Perceptron Implementation from Scratch",
 229 |         "day_break":false,
 230 |         "book":"chapter_multilayer-perceptrons/mlp-scratch.html",
 231 |         "slides":["part-0.pdf",0],
 232 |         "slides_video":"",
 233 |         "notebook_video":"",
 234 |         "qa_video":""
 235 |     },
 236 |     {
 237 |         "title":"Concise Implementation of Multilayer Perceptrons",
 238 |         "day_break":false,
 239 |         "book":"chapter_multilayer-perceptrons/mlp-concise.html",
 240 |         "slides":["part-0.pdf",0],
 241 |         "slides_video":"",
 242 |         "notebook_video":"",
 243 |         "qa_video":""
 244 |     },
 245 |     {
 246 |         "title":"Model Selection, Underfitting, and Overfitting",
 247 |         "day_break":false,
 248 |         "book":"chapter_multilayer-perceptrons/underfit-overfit.html",
 249 |         "slides":["part-0.pdf",0],
 250 |         "slides_video":"",
 251 |         "notebook_video":"",
 252 |         "qa_video":""
 253 |     },
 254 |     {
 255 |         "title":"Weight Decay",
 256 |         "day_break":false,
 257 |         "book":"chapter_multilayer-perceptrons/weight-decay.html",
 258 |         "slides":["part-0.pdf",0],
 259 |         "slides_video":"",
 260 |         "notebook_video":"",
 261 |         "qa_video":""
 262 |     },
 263 |     {
 264 |         "title":"Dropout",
 265 |         "day_break":false,
 266 |         "book":"chapter_multilayer-perceptrons/dropout.html",
 267 |         "slides":["part-0.pdf",0],
 268 |         "slides_video":"",
 269 |         "notebook_video":"",
 270 |         "qa_video":""
 271 |     },
 272 |     {
 273 |         "title":"Forward Propagation, Backward Propagation, and Computational Graphs",
 274 |         "day_break":false,
 275 |         "book":"chapter_multilayer-perceptrons/backprop.html",
 276 |         "slides":["part-0.pdf",0],
 277 |         "slides_video":"",
 278 |         "notebook_video":"",
 279 |         "qa_video":""
 280 |     },
 281 |     {
 282 |         "title":"Numerical Stability and Initialization",
 283 |         "day_break":false,
 284 |         "book":"chapter_multilayer-perceptrons/numerical-stability-and-init.html",
 285 |         "slides":["part-0.pdf",0],
 286 |         "slides_video":"",
 287 |         "notebook_video":"",
 288 |         "qa_video":""
 289 |     },
 290 |     {
 291 |         "title":"Environment and Distribution Shift",
 292 |         "day_break":false,
 293 |         "book":"chapter_multilayer-perceptrons/environment.html",
 294 |         "slides":["part-0.pdf",0],
 295 |         "slides_video":"",
 296 |         "notebook_video":"",
 297 |         "qa_video":""
 298 |     },
 299 |     {
 300 |         "title":"Predicting House Prices on Kaggle",
 301 |         "day_break":false,
 302 |         "book":"chapter_multilayer-perceptrons/kaggle-house-price.html",
 303 |         "slides":["part-0.pdf",0],
 304 |         "slides_video":"",
 305 |         "notebook_video":"",
 306 |         "qa_video":""
 307 |     },
 308 |     {
 309 |         "title":"Deep Learning Computation",
 310 |         "day_break":false,
 311 |         "book":"chapter_deep-learning-computation/index.html",
 312 |         "slides":["part-0.pdf",0],
 313 |         "slides_video":"",
 314 |         "notebook_video":"",
 315 |         "qa_video":""
 316 |     },
 317 |     {
 318 |         "title":"Layers and Blocks",
 319 |         "day_break":false,
 320 |         "book":"chapter_deep-learning-computation/model-construction.html",
 321 |         "slides":["part-0.pdf",0],
 322 |         "slides_video":"",
 323 |         "notebook_video":"",
 324 |         "qa_video":""
 325 |     },
 326 |     {
 327 |         "title":"Parameter Management",
 328 |         "day_break":false,
 329 |         "book":"chapter_deep-learning-computation/parameters.html",
 330 |         "slides":["part-0.pdf",0],
 331 |         "slides_video":"",
 332 |         "notebook_video":"",
 333 |         "qa_video":""
 334 |     },
 335 |     {
 336 |         "title":"Deferred Initialization",
 337 |         "day_break":false,
 338 |         "book":"chapter_deep-learning-computation/deferred-init.html",
 339 |         "slides":["part-0.pdf",0],
 340 |         "slides_video":"",
 341 |         "notebook_video":"",
 342 |         "qa_video":""
 343 |     },
 344 |     {
 345 |         "title":"Custom Layers",
 346 |         "day_break":false,
 347 |         "book":"chapter_deep-learning-computation/custom-layer.html",
 348 |         "slides":["part-0.pdf",0],
 349 |         "slides_video":"",
 350 |         "notebook_video":"",
 351 |         "qa_video":""
 352 |     },
 353 |     {
 354 |         "title":"File I/O",
 355 |         "day_break":false,
 356 |         "book":"chapter_deep-learning-computation/read-write.html",
 357 |         "slides":["part-0.pdf",0],
 358 |         "slides_video":"",
 359 |         "notebook_video":"",
 360 |         "qa_video":""
 361 |     },
 362 |     {
 363 |         "title":"GPUs",
 364 |         "day_break":false,
 365 |         "book":"chapter_deep-learning-computation/use-gpu.html",
 366 |         "slides":["part-0.pdf",0],
 367 |         "slides_video":"",
 368 |         "notebook_video":"",
 369 |         "qa_video":""
 370 |     },
 371 |     {
 372 |         "title":"Convolutional Neural Networks",
 373 |         "day_break":false,
 374 |         "book":"chapter_convolutional-neural-networks/index.html",
 375 |         "slides":["part-0.pdf",0],
 376 |         "slides_video":"",
 377 |         "notebook_video":"",
 378 |         "qa_video":""
 379 |     },
 380 |     {
 381 |         "title":"From Fully Connected Layers to Convolutions",
 382 |         "day_break":false,
 383 |         "book":"chapter_convolutional-neural-networks/why-conv.html",
 384 |         "slides":["part-0.pdf",0],
 385 |         "slides_video":"",
 386 |         "notebook_video":"",
 387 |         "qa_video":""
 388 |     },
 389 |     {
 390 |         "title":"Convolutions for Images",
 391 |         "day_break":false,
 392 |         "book":"chapter_convolutional-neural-networks/conv-layer.html",
 393 |         "slides":["part-0.pdf",0],
 394 |         "slides_video":"",
 395 |         "notebook_video":"",
 396 |         "qa_video":""
 397 |     },
 398 |     {
 399 |         "title":"Padding and Stride",
 400 |         "day_break":false,
 401 |         "book":"chapter_convolutional-neural-networks/padding-and-strides.html",
 402 |         "slides":["part-0.pdf",0],
 403 |         "slides_video":"",
 404 |         "notebook_video":"",
 405 |         "qa_video":""
 406 |     },
 407 |     {
 408 |         "title":"Multiple Input and Multiple Output Channels",
 409 |         "day_break":false,
 410 |         "book":"chapter_convolutional-neural-networks/channels.html",
 411 |         "slides":["part-0.pdf",0],
 412 |         "slides_video":"",
 413 |         "notebook_video":"",
 414 |         "qa_video":""
 415 |     },
 416 |     {
 417 |         "title":"Pooling",
 418 |         "day_break":false,
 419 |         "book":"chapter_convolutional-neural-networks/pooling.html",
 420 |         "slides":["part-0.pdf",0],
 421 |         "slides_video":"",
 422 |         "notebook_video":"",
 423 |         "qa_video":""
 424 |     },
 425 |     {
 426 |         "title":"Convolutional Neural Networks (LeNet)",
 427 |         "day_break":false,
 428 |         "book":"chapter_convolutional-neural-networks/lenet.html",
 429 |         "slides":["part-0.pdf",0],
 430 |         "slides_video":"",
 431 |         "notebook_video":"",
 432 |         "qa_video":""
 433 |     },
 434 |     {
 435 |         "title":"Modern Convolutional Neural Networks",
 436 |         "day_break":false,
 437 |         "book":"chapter_convolutional-modern/index.html",
 438 |         "slides":["part-0.pdf",0],
 439 |         "slides_video":"",
 440 |         "notebook_video":"",
 441 |         "qa_video":""
 442 |     },
 443 |     {
 444 |         "title":"Deep Convolutional Neural Networks (AlexNet)",
 445 |         "day_break":false,
 446 |         "book":"chapter_convolutional-modern/alexnet.html",
 447 |         "slides":["part-0.pdf",0],
 448 |         "slides_video":"",
 449 |         "notebook_video":"",
 450 |         "qa_video":""
 451 |     },
 452 |     {
 453 |         "title":"Networks Using Blocks (VGG)",
 454 |         "day_break":false,
 455 |         "book":"chapter_convolutional-modern/vgg.html",
 456 |         "slides":["part-0.pdf",0],
 457 |         "slides_video":"",
 458 |         "notebook_video":"",
 459 |         "qa_video":""
 460 |     },
 461 |     {
 462 |         "title":"Network in Network (NiN)",
 463 |         "day_break":false,
 464 |         "book":"chapter_convolutional-modern/nin.html",
 465 |         "slides":["part-0.pdf",0],
 466 |         "slides_video":"",
 467 |         "notebook_video":"",
 468 |         "qa_video":""
 469 |     },
 470 |     {
 471 |         "title":"Networks with Parallel Concatenations (GoogLeNet)",
 472 |         "day_break":false,
 473 |         "book":"chapter_convolutional-modern/googlenet.html",
 474 |         "slides":["part-0.pdf",0],
 475 |         "slides_video":"",
 476 |         "notebook_video":"",
 477 |         "qa_video":""
 478 |     },
 479 |     {
 480 |         "title":"Batch Normalization",
 481 |         "day_break":false,
 482 |         "book":"chapter_convolutional-modern/batch-norm.html",
 483 |         "slides":["part-0.pdf",0],
 484 |         "slides_video":"",
 485 |         "notebook_video":"",
 486 |         "qa_video":""
 487 |     },
 488 |     {
 489 |         "title":"Residual Networks (ResNet)",
 490 |         "day_break":false,
 491 |         "book":"chapter_convolutional-modern/resnet.html",
 492 |         "slides":["part-0.pdf",0],
 493 |         "slides_video":"",
 494 |         "notebook_video":"",
 495 |         "qa_video":""
 496 |     },
 497 |     {
 498 |         "title":"Densely Connected Networks (DenseNet)",
 499 |         "day_break":false,
 500 |         "book":"chapter_convolutional-modern/densenet.html",
 501 |         "slides":["part-0.pdf",0],
 502 |         "slides_video":"",
 503 |         "notebook_video":"",
 504 |         "qa_video":""
 505 |     },
 506 |     {
 507 |         "title":"Recurrent Neural Networks",
 508 |         "day_break":false,
 509 |         "book":"chapter_recurrent-neural-networks/index.html",
 510 |         "slides":["part-0.pdf",0],
 511 |         "slides_video":"",
 512 |         "notebook_video":"",
 513 |         "qa_video":""
 514 |     },
 515 |     {
 516 |         "title":"Sequence Models",
 517 |         "day_break":false,
 518 |         "book":"chapter_recurrent-neural-networks/sequence.html",
 519 |         "slides":["part-0.pdf",0],
 520 |         "slides_video":"",
 521 |         "notebook_video":"",
 522 |         "qa_video":""
 523 |     },
 524 |     {
 525 |         "title":"Text Preprocessing",
 526 |         "day_break":false,
 527 |         "book":"chapter_recurrent-neural-networks/text-preprocessing.html",
 528 |         "slides":["part-0.pdf",0],
 529 |         "slides_video":"",
 530 |         "notebook_video":"",
 531 |         "qa_video":""
 532 |     },
 533 |     {
 534 |         "title":"The Language Model Dataset",
 535 |         "day_break":false,
 536 |         "book":"chapter_recurrent-neural-networks/language-models-and-dataset.html",
 537 |         "slides":["part-0.pdf",0],
 538 |         "slides_video":"",
 539 |         "notebook_video":"",
 540 |         "qa_video":""
 541 |     },
 542 |     {
 543 |         "title":"Recurrent Neural Networks",
 544 |         "day_break":false,
 545 |         "book":"chapter_recurrent-neural-networks/rnn.html",
 546 |         "slides":["part-0.pdf",0],
 547 |         "slides_video":"",
 548 |         "notebook_video":"",
 549 |         "qa_video":""
 550 |     },
 551 |     {
 552 |         "title":"Recurrent Neural Network Implementation from Scratch",
 553 |         "day_break":false,
 554 |         "book":"chapter_recurrent-neural-networks/rnn-scratch.html",
 555 |         "slides":["part-0.pdf",0],
 556 |         "slides_video":"",
 557 |         "notebook_video":"",
 558 |         "qa_video":""
 559 |     },
 560 |     {
 561 |         "title":"Concise Implementation of Recurrent Neural Networks",
 562 |         "day_break":false,
 563 |         "book":"chapter_recurrent-neural-networks/rnn-concise.html",
 564 |         "slides":["part-0.pdf",0],
 565 |         "slides_video":"",
 566 |         "notebook_video":"",
 567 |         "qa_video":""
 568 |     },
 569 |     {
 570 |         "title":"Backpropagation Through Time",
 571 |         "day_break":false,
 572 |         "book":"chapter_recurrent-neural-networks/bptt.html",
 573 |         "slides":["part-0.pdf",0],
 574 |         "slides_video":"",
 575 |         "notebook_video":"",
 576 |         "qa_video":""
 577 |     },
 578 |     {
 579 |         "title":"Modern Recurrent Neural Networks",
 580 |         "day_break":false,
 581 |         "book":"chapter_recurrent-modern/index.html",
 582 |         "slides":["part-0.pdf",0],
 583 |         "slides_video":"",
 584 |         "notebook_video":"",
 585 |         "qa_video":""
 586 |     },
 587 |     {
 588 |         "title":"Gated Recurrent Units (GRU)",
 589 |         "day_break":false,
 590 |         "book":"chapter_recurrent-modern/gru.html",
 591 |         "slides":["part-0.pdf",0],
 592 |         "slides_video":"",
 593 |         "notebook_video":"",
 594 |         "qa_video":""
 595 |     },
 596 |     {
 597 |         "title":"Long Short-Term Memory (LSTM)",
 598 |         "day_break":false,
 599 |         "book":"chapter_recurrent-modern/lstm.html",
 600 |         "slides":["part-0.pdf",0],
 601 |         "slides_video":"",
 602 |         "notebook_video":"",
 603 |         "qa_video":""
 604 |     },
 605 |     {
 606 |         "title":"Deep Recurrent Neural Networks",
 607 |         "day_break":false,
 608 |         "book":"chapter_recurrent-modern/deep-rnn.html",
 609 |         "slides":["part-0.pdf",0],
 610 |         "slides_video":"",
 611 |         "notebook_video":"",
 612 |         "qa_video":""
 613 |     },
 614 |     {
 615 |         "title":"Bidirectional Recurrent Neural Networks",
 616 |         "day_break":false,
 617 |         "book":"chapter_recurrent-modern/bi-rnn.html",
 618 |         "slides":["part-0.pdf",0],
 619 |         "slides_video":"",
 620 |         "notebook_video":"",
 621 |         "qa_video":""
 622 |     },
 623 |     {
 624 |         "title":"Machine Translation and the Dataset",
 625 |         "day_break":false,
 626 |         "book":"chapter_recurrent-modern/machine-translation-and-dataset.html",
 627 |         "slides":["part-0.pdf",0],
 628 |         "slides_video":"",
 629 |         "notebook_video":"",
 630 |         "qa_video":""
 631 |     },
 632 |     {
 633 |         "title":"Encoder-Decoder Architecture",
 634 |         "day_break":false,
 635 |         "book":"chapter_recurrent-modern/encoder-decoder.html",
 636 |         "slides":["part-0.pdf",0],
 637 |         "slides_video":"",
 638 |         "notebook_video":"",
 639 |         "qa_video":""
 640 |     },
 641 |     {
 642 |         "title":"Sequence to Sequence Learning",
 643 |         "day_break":false,
 644 |         "book":"chapter_recurrent-modern/seq2seq.html",
 645 |         "slides":["part-0.pdf",0],
 646 |         "slides_video":"",
 647 |         "notebook_video":"",
 648 |         "qa_video":""
 649 |     },
 650 |     {
 651 |         "title":"Beam Search",
 652 |         "day_break":false,
 653 |         "book":"chapter_recurrent-modern/beam-search.html",
 654 |         "slides":["part-0.pdf",0],
 655 |         "slides_video":"",
 656 |         "notebook_video":"",
 657 |         "qa_video":""
 658 |     },
 659 |     {
 660 |         "title":"Attention Mechanisms",
 661 |         "day_break":false,
 662 |         "book":"chapter_attention-mechanisms/index.html",
 663 |         "slides":["part-0.pdf",0],
 664 |         "slides_video":"",
 665 |         "notebook_video":"",
 666 |         "qa_video":""
 667 |     },
 668 |     {
 669 |         "title":"Attention Cues",
 670 |         "day_break":false,
 671 |         "book":"chapter_attention-mechanisms/attention-cues.html",
 672 |         "slides":["part-0.pdf",0],
 673 |         "slides_video":"",
 674 |         "notebook_video":"",
 675 |         "qa_video":""
 676 |     },
 677 |     {
 678 |         "title":"Attention Pooling: Nadaraya-Watson Kernel Regression",
 679 |         "day_break":false,
 680 |         "book":"chapter_attention-mechanisms/nadaraya-watson.html",
 681 |         "slides":["part-0.pdf",0],
 682 |         "slides_video":"",
 683 |         "notebook_video":"",
 684 |         "qa_video":""
 685 |     },
 686 |     {
 687 |         "title":"Attention Scoring Functions",
 688 |         "day_break":false,
 689 |         "book":"chapter_attention-mechanisms/attention-scoring-functions.html",
 690 |         "slides":["part-0.pdf",0],
 691 |         "slides_video":"",
 692 |         "notebook_video":"",
 693 |         "qa_video":""
 694 |     },
 695 |     {
 696 |         "title":"Bahdanau Attention",
 697 |         "day_break":false,
 698 |         "book":"chapter_attention-mechanisms/bahdanau-attention.html",
 699 |         "slides":["part-0.pdf",0],
 700 |         "slides_video":"",
 701 |         "notebook_video":"",
 702 |         "qa_video":""
 703 |     },
 704 |     {
 705 |         "title":"Multi-Head Attention",
 706 |         "day_break":false,
 707 |         "book":"chapter_attention-mechanisms/multihead-attention.html",
 708 |         "slides":["part-0.pdf",0],
 709 |         "slides_video":"",
 710 |         "notebook_video":"",
 711 |         "qa_video":""
 712 |     },
 713 |     {
 714 |         "title":"Self-Attention and Positional Encoding",
 715 |         "day_break":false,
 716 |         "book":"chapter_attention-mechanisms/self-attention-and-positional-encoding.html",
 717 |         "slides":["part-0.pdf",0],
 718 |         "slides_video":"",
 719 |         "notebook_video":"",
 720 |         "qa_video":""
 721 |     },
 722 |     {
 723 |         "title":"Transformer",
 724 |         "day_break":false,
 725 |         "book":"chapter_attention-mechanisms/transformer.html",
 726 |         "slides":["part-0.pdf",0],
 727 |         "slides_video":"",
 728 |         "notebook_video":"",
 729 |         "qa_video":""
 730 |     },
 731 |     {
 732 |         "title":"Optimization Algorithms",
 733 |         "day_break":false,
 734 |         "book":"chapter_optimization/index.html",
 735 |         "slides":["part-0.pdf",0],
 736 |         "slides_video":"",
 737 |         "notebook_video":"",
 738 |         "qa_video":""
 739 |     },
 740 |     {
 741 |         "title":"Optimization and Deep Learning",
 742 |         "day_break":false,
 743 |         "book":"chapter_optimization/optimization-intro.html",
 744 |         "slides":["part-0.pdf",0],
 745 |         "slides_video":"",
 746 |         "notebook_video":"",
 747 |         "qa_video":""
 748 |     },
 749 |     {
 750 |         "title":"Convexity",
 751 |         "day_break":false,
 752 |         "book":"chapter_optimization/convexity.html",
 753 |         "slides":["part-0.pdf",0],
 754 |         "slides_video":"",
 755 |         "notebook_video":"",
 756 |         "qa_video":""
 757 |     },
 758 |     {
 759 |         "title":"Gradient Descent",
 760 |         "day_break":false,
 761 |         "book":"chapter_optimization/gd.html",
 762 |         "slides":["part-0.pdf",0],
 763 |         "slides_video":"",
 764 |         "notebook_video":"",
 765 |         "qa_video":""
 766 |     },
 767 |     {
 768 |         "title":"Stochastic Gradient Descent",
 769 |         "day_break":false,
 770 |         "book":"chapter_optimization/sgd.html",
 771 |         "slides":["part-0.pdf",0],
 772 |         "slides_video":"",
 773 |         "notebook_video":"",
 774 |         "qa_video":""
 775 |     },
 776 |     {
 777 |         "title":"Minibatch Stochastic Gradient Descent",
 778 |         "day_break":false,
 779 |         "book":"chapter_optimization/minibatch-sgd.html",
 780 |         "slides":["part-0.pdf",0],
 781 |         "slides_video":"",
 782 |         "notebook_video":"",
 783 |         "qa_video":""
 784 |     },
 785 |     {
 786 |         "title":"Momentum",
 787 |         "day_break":false,
 788 |         "book":"chapter_optimization/momentum.html",
 789 |         "slides":["part-0.pdf",0],
 790 |         "slides_video":"",
 791 |         "notebook_video":"",
 792 |         "qa_video":""
 793 |     },
 794 |     {
 795 |         "title":"Adagrad",
 796 |         "day_break":false,
 797 |         "book":"chapter_optimization/adagrad.html",
 798 |         "slides":["part-0.pdf",0],
 799 |         "slides_video":"",
 800 |         "notebook_video":"",
 801 |         "qa_video":""
 802 |     },
 803 |     {
 804 |         "title":"RMSProp",
 805 |         "day_break":false,
 806 |         "book":"chapter_optimization/rmsprop.html",
 807 |         "slides":["part-0.pdf",0],
 808 |         "slides_video":"",
 809 |         "notebook_video":"",
 810 |         "qa_video":""
 811 |     },
 812 |     {
 813 |         "title":"Adadelta",
 814 |         "day_break":false,
 815 |         "book":"chapter_optimization/adadelta.html",
 816 |         "slides":["part-0.pdf",0],
 817 |         "slides_video":"",
 818 |         "notebook_video":"",
 819 |         "qa_video":""
 820 |     },
 821 |     {
 822 |         "title":"Adam",
 823 |         "day_break":false,
 824 |         "book":"chapter_optimization/adam.html",
 825 |         "slides":["part-0.pdf",0],
 826 |         "slides_video":"",
 827 |         "notebook_video":"",
 828 |         "qa_video":""
 829 |     },
 830 |     {
 831 |         "title":"Learning Rate Scheduling",
 832 |         "day_break":false,
 833 |         "book":"chapter_optimization/lr-scheduler.html",
 834 |         "slides":["part-0.pdf",0],
 835 |         "slides_video":"",
 836 |         "notebook_video":"",
 837 |         "qa_video":""
 838 |     },
 839 |     {
 840 |         "title":"Computational Performance",
 841 |         "day_break":false,
 842 |         "book":"chapter_computational-performance/index.html",
 843 |         "slides":["part-0.pdf",0],
 844 |         "slides_video":"",
 845 |         "notebook_video":"",
 846 |         "qa_video":""
 847 |     },
 848 |     {
 849 |         "title":"Compilers and Interpreters",
 850 |         "day_break":false,
 851 |         "book":"chapter_computational-performance/hybridize.html",
 852 |         "slides":["part-0.pdf",0],
 853 |         "slides_video":"",
 854 |         "notebook_video":"",
 855 |         "qa_video":""
 856 |     },
 857 |     {
 858 |         "title":"Asynchronous Computation",
 859 |         "day_break":false,
 860 |         "book":"chapter_computational-performance/async-computation.html",
 861 |         "slides":["part-0.pdf",0],
 862 |         "slides_video":"",
 863 |         "notebook_video":"",
 864 |         "qa_video":""
 865 |     },
 866 |     {
 867 |         "title":"Automatic Parallelism",
 868 |         "day_break":false,
 869 |         "book":"chapter_computational-performance/auto-parallelism.html",
 870 |         "slides":["part-0.pdf",0],
 871 |         "slides_video":"",
 872 |         "notebook_video":"",
 873 |         "qa_video":""
 874 |     },
 875 |     {
 876 |         "title":"Hardware",
 877 |         "day_break":false,
 878 |         "book":"chapter_computational-performance/hardware.html",
 879 |         "slides":["part-0.pdf",0],
 880 |         "slides_video":"",
 881 |         "notebook_video":"",
 882 |         "qa_video":""
 883 |     },
 884 |     {
 885 |         "title":"Training on Multiple GPUs",
 886 |         "day_break":false,
 887 |         "book":"chapter_computational-performance/multiple-gpus.html",
 888 |         "slides":["part-0.pdf",0],
 889 |         "slides_video":"",
 890 |         "notebook_video":"",
 891 |         "qa_video":""
 892 |     },
 893 |     {
 894 |         "title":"Concise Implementation for Multiple GPUs",
 895 |         "day_break":false,
 896 |         "book":"chapter_computational-performance/multiple-gpus-concise.html",
 897 |         "slides":["part-0.pdf",0],
 898 |         "slides_video":"",
 899 |         "notebook_video":"",
 900 |         "qa_video":""
 901 |     },
 902 |     {
 903 |         "title":"Parameter Servers",
 904 |         "day_break":false,
 905 |         "book":"chapter_computational-performance/parameterserver.html",
 906 |         "slides":["part-0.pdf",0],
 907 |         "slides_video":"",
 908 |         "notebook_video":"",
 909 |         "qa_video":""
 910 |     },
 911 |     {
 912 |         "title":"Computer Vision",
 913 |         "day_break":false,
 914 |         "book":"chapter_computer-vision/index.html",
 915 |         "slides":["part-0.pdf",0],
 916 |         "slides_video":"",
 917 |         "notebook_video":"",
 918 |         "qa_video":""
 919 |     },
 920 |     {
 921 |         "title":"Image Augmentation",
 922 |         "day_break":false,
 923 |         "book":"chapter_computer-vision/image-augmentation.html",
 924 |         "slides":["part-0.pdf",0],
 925 |         "slides_video":"",
 926 |         "notebook_video":"",
 927 |         "qa_video":""
 928 |     },
 929 |     {
 930 |         "title":"Fine-Tuning",
 931 |         "day_break":false,
 932 |         "book":"chapter_computer-vision/fine-tuning.html",
 933 |         "slides":["part-0.pdf",0],
 934 |         "slides_video":"",
 935 |         "notebook_video":"",
 936 |         "qa_video":""
 937 |     },
 938 |     {
 939 |         "title":"Object Detection and Bounding Boxes",
 940 |         "day_break":false,
 941 |         "book":"chapter_computer-vision/bounding-box.html",
 942 |         "slides":["part-0.pdf",0],
 943 |         "slides_video":"",
 944 |         "notebook_video":"",
 945 |         "qa_video":""
 946 |     },
 947 |     {
 948 |         "title":"Anchor Boxes",
 949 |         "day_break":false,
 950 |         "book":"chapter_computer-vision/anchor.html",
 951 |         "slides":["part-0.pdf",0],
 952 |         "slides_video":"",
 953 |         "notebook_video":"",
 954 |         "qa_video":""
 955 |     },
 956 |     {
 957 |         "title":"Multiscale Object Detection",
 958 |         "day_break":false,
 959 |         "book":"chapter_computer-vision/multiscale-object-detection.html",
 960 |         "slides":["part-0.pdf",0],
 961 |         "slides_video":"",
 962 |         "notebook_video":"",
 963 |         "qa_video":""
 964 |     },
 965 |     {
 966 |         "title":"The Object Detection Dataset",
 967 |         "day_break":false,
 968 |         "book":"chapter_computer-vision/object-detection-dataset.html",
 969 |         "slides":["part-0.pdf",0],
 970 |         "slides_video":"",
 971 |         "notebook_video":"",
 972 |         "qa_video":""
 973 |     },
 974 |     {
 975 |         "title":"Single Shot Multibox Detection",
 976 |         "day_break":false,
 977 |         "book":"chapter_computer-vision/ssd.html",
 978 |         "slides":["part-0.pdf",0],
 979 |         "slides_video":"",
 980 |         "notebook_video":"",
 981 |         "qa_video":""
 982 |     },
 983 |     {
 984 |         "title":"Region-based CNNs (R-CNNs)",
 985 |         "day_break":false,
 986 |         "book":"chapter_computer-vision/rcnn.html",
 987 |         "slides":["part-0.pdf",0],
 988 |         "slides_video":"",
 989 |         "notebook_video":"",
 990 |         "qa_video":""
 991 |     },
 992 |     {
 993 |         "title":"Semantic Segmentation and the Dataset",
 994 |         "day_break":false,
 995 |         "book":"chapter_computer-vision/semantic-segmentation-and-dataset.html",
 996 |         "slides":["part-0.pdf",0],
 997 |         "slides_video":"",
 998 |         "notebook_video":"",
 999 |         "qa_video":""
1000 |     },
1001 |     {
1002 |         "title":"Transposed Convolution",
1003 |         "day_break":false,
1004 |         "book":"chapter_computer-vision/transposed-conv.html",
1005 |         "slides":["part-0.pdf",0],
1006 |         "slides_video":"",
1007 |         "notebook_video":"",
1008 |         "qa_video":""
1009 |     },
1010 |     {
1011 |         "title":"Fully Convolutional Networks",
1012 |         "day_break":false,
1013 |         "book":"chapter_computer-vision/fcn.html",
1014 |         "slides":["part-0.pdf",0],
1015 |         "slides_video":"",
1016 |         "notebook_video":"",
1017 |         "qa_video":""
1018 |     },
1019 |     {
1020 |         "title":"Neural Style Transfer",
1021 |         "day_break":false,
1022 |         "book":"chapter_computer-vision/neural-style.html",
1023 |         "slides":["part-0.pdf",0],
1024 |         "slides_video":"",
1025 |         "notebook_video":"",
1026 |         "qa_video":""
1027 |     },
1028 |     {
1029 |         "title":"Image Classification (CIFAR-10) on Kaggle",
1030 |         "day_break":false,
1031 |         "book":"chapter_computer-vision/kaggle-cifar10.html",
1032 |         "slides":["part-0.pdf",0],
1033 |         "slides_video":"",
1034 |         "notebook_video":"",
1035 |         "qa_video":""
1036 |     },
1037 |     {
1038 |         "title":"Dog Breed Identification (ImageNet Dogs) on Kaggle",
1039 |         "day_break":false,
1040 |         "book":"chapter_computer-vision/kaggle-dog.html",
1041 |         "slides":["part-0.pdf",0],
1042 |         "slides_video":"",
1043 |         "notebook_video":"",
1044 |         "qa_video":""
1045 |     },
1046 |     {
1047 |         "title":"Natural Language Processing: Pretraining",
1048 |         "day_break":false,
1049 |         "book":"chapter_natural-language-processing-pretraining/index.html",
1050 |         "slides":["part-0.pdf",0],
1051 |         "slides_video":"",
1052 |         "notebook_video":"",
1053 |         "qa_video":""
1054 |     },
1055 |     {
1056 |         "title":"Word Embedding (word2vec)",
1057 |         "day_break":false,
1058 |         "book":"chapter_natural-language-processing-pretraining/word2vec.html",
1059 |         "slides":["part-0.pdf",0],
1060 |         "slides_video":"",
1061 |         "notebook_video":"",
1062 |         "qa_video":""
1063 |     },
1064 |     {
1065 |         "title":"Approximate Training",
1066 |         "day_break":false,
1067 |         "book":"chapter_natural-language-processing-pretraining/approx-training.html",
1068 |         "slides":["part-0.pdf",0],
1069 |         "slides_video":"",
1070 |         "notebook_video":"",
1071 |         "qa_video":""
1072 |     },
1073 |     {
1074 |         "title":"The Dataset for Pretraining Word Embeddings",
1075 |         "day_break":false,
1076 |         "book":"chapter_natural-language-processing-pretraining/word-embedding-dataset.html",
1077 |         "slides":["part-0.pdf",0],
1078 |         "slides_video":"",
1079 |         "notebook_video":"",
1080 |         "qa_video":""
1081 |     },
1082 |     {
1083 |         "title":"Pretraining word2vec",
1084 |         "day_break":false,
1085 |         "book":"chapter_natural-language-processing-pretraining/word2vec-pretraining.html",
1086 |         "slides":["part-0.pdf",0],
1087 |         "slides_video":"",
1088 |         "notebook_video":"",
1089 |         "qa_video":""
1090 |     },
1091 |     {
1092 |         "title":"Word Embedding with Global Vectors (GloVe)",
1093 |         "day_break":false,
1094 |         "book":"chapter_natural-language-processing-pretraining/glove.html",
1095 |         "slides":["part-0.pdf",0],
1096 |         "slides_video":"",
1097 |         "notebook_video":"",
1098 |         "qa_video":""
1099 |     },
1100 |     {
1101 |         "title":"Subword Embedding",
1102 |         "day_break":false,
1103 |         "book":"chapter_natural-language-processing-pretraining/subword-embedding.html",
1104 |         "slides":["part-0.pdf",0],
1105 |         "slides_video":"",
1106 |         "notebook_video":"",
1107 |         "qa_video":""
1108 |     },
1109 |     {
1110 |         "title":"Word Similarity and Analogy",
1111 |         "day_break":false,
1112 |         "book":"chapter_natural-language-processing-pretraining/similarity-analogy.html",
1113 |         "slides":["part-0.pdf",0],
1114 |         "slides_video":"",
1115 |         "notebook_video":"",
1116 |         "qa_video":""
1117 |     },
1118 |     {
1119 |         "title":"Bidirectional Encoder Representations from Transformers (BERT)",
1120 |         "day_break":false,
1121 |         "book":"chapter_natural-language-processing-pretraining/bert.html",
1122 |         "slides":["part-0.pdf",0],
1123 |         "slides_video":"",
1124 |         "notebook_video":"",
1125 |         "qa_video":""
1126 |     },
1127 |     {
1128 |         "title":"The Dataset for Pretraining BERT",
1129 |         "day_break":false,
1130 |         "book":"chapter_natural-language-processing-pretraining/bert-dataset.html",
1131 |         "slides":["part-0.pdf",0],
1132 |         "slides_video":"",
1133 |         "notebook_video":"",
1134 |         "qa_video":""
1135 |     },
1136 |     {
1137 |         "title":"Pretraining BERT",
1138 |         "day_break":false,
1139 |         "book":"chapter_natural-language-processing-pretraining/bert-pretraining.html",
1140 |         "slides":["part-0.pdf",0],
1141 |         "slides_video":"",
1142 |         "notebook_video":"",
1143 |         "qa_video":""
1144 |     },
1145 |     {
1146 |         "title":"Natural Language Processing: Applications",
1147 |         "day_break":false,
1148 |         "book":"chapter_natural-language-processing-applications/index.html",
1149 |         "slides":["part-0.pdf",0],
1150 |         "slides_video":"",
1151 |         "notebook_video":"",
1152 |         "qa_video":""
1153 |     },
1154 |     {
1155 |         "title":"Sentiment Analysis and the Dataset",
1156 |         "day_break":false,
1157 |         "book":"chapter_natural-language-processing-applications/sentiment-analysis-and-dataset.html",
1158 |         "slides":["part-0.pdf",0],
1159 |         "slides_video":"",
1160 |         "notebook_video":"",
1161 |         "qa_video":""
1162 |     },
1163 |     {
1164 |         "title":"Sentiment Analysis: Using Recurrent Neural Networks",
1165 |         "day_break":false,
1166 |         "book":"chapter_natural-language-processing-applications/sentiment-analysis-rnn.html",
1167 |         "slides":["part-0.pdf",0],
1168 |         "slides_video":"",
1169 |         "notebook_video":"",
1170 |         "qa_video":""
1171 |     },
1172 |     {
1173 |         "title":"Sentiment Analysis: Using Convolutional Neural Networks",
1174 |         "day_break":false,
1175 |         "book":"chapter_natural-language-processing-applications/sentiment-analysis-cnn.html",
1176 |         "slides":["part-0.pdf",0],
1177 |         "slides_video":"",
1178 |         "notebook_video":"",
1179 |         "qa_video":""
1180 |     },
1181 |     {
1182 |         "title":"Natural Language Inference and the Dataset",
1183 |         "day_break":false,
1184 |         "book":"chapter_natural-language-processing-applications/natural-language-inference-and-dataset.html",
1185 |         "slides":["part-0.pdf",0],
1186 |         "slides_video":"",
1187 |         "notebook_video":"",
1188 |         "qa_video":""
1189 |     },
1190 |     {
1191 |         "title":"Natural Language Inference: Using Attention",
1192 |         "day_break":false,
1193 |         "book":"chapter_natural-language-processing-applications/natural-language-inference-attention.html",
1194 |         "slides":["part-0.pdf",0],
1195 |         "slides_video":"",
1196 |         "notebook_video":"",
1197 |         "qa_video":""
1198 |     },
1199 |     {
1200 |         "title":"Fine-Tuning BERT for Sequence-Level and Token-Level Applications",
1201 |         "day_break":false,
1202 |         "book":"chapter_natural-language-processing-applications/finetuning-bert.html",
1203 |         "slides":["part-0.pdf",0],
1204 |         "slides_video":"",
1205 |         "notebook_video":"",
1206 |         "qa_video":""
1207 |     },
1208 |     {
1209 |         "title":"Natural Language Inference: Fine-Tuning BERT",
1210 |         "day_break":false,
1211 |         "book":"chapter_natural-language-processing-applications/natural-language-inference-bert.html",
1212 |         "slides":["part-0.pdf",0],
1213 |         "slides_video":"",
1214 |         "notebook_video":"",
1215 |         "qa_video":""
1216 |     },
1217 |     {
1218 |         "title":"Recommender Systems",
1219 |         "day_break":false,
1220 |         "book":"chapter_recommender-systems/index.html",
1221 |         "slides":["part-0.pdf",0],
1222 |         "slides_video":"",
1223 |         "notebook_video":"",
1224 |         "qa_video":""
1225 |     },
1226 |     {
1227 |         "title":"Overview of Recommender Systems",
1228 |         "day_break":false,
1229 |         "book":"chapter_recommender-systems/recsys-intro.html",
1230 |         "slides":["part-0.pdf",0],
1231 |         "slides_video":"",
1232 |         "notebook_video":"",
1233 |         "qa_video":""
1234 |     },
1235 |     {
1236 |         "title":"The MovieLens Dataset",
1237 |         "day_break":false,
1238 |         "book":"chapter_recommender-systems/movielens.html",
1239 |         "slides":["part-0.pdf",0],
1240 |         "slides_video":"",
1241 |         "notebook_video":"",
1242 |         "qa_video":""
1243 |     },
1244 |     {
1245 |         "title":"Matrix Factorization",
1246 |         "day_break":false,
1247 |         "book":"chapter_recommender-systems/mf.html",
1248 |         "slides":["part-0.pdf",0],
1249 |         "slides_video":"",
1250 |         "notebook_video":"",
1251 |         "qa_video":""
1252 |     },
1253 |     {
1254 |         "title":"AutoRec: Rating Prediction with Autoencoders",
1255 |         "day_break":false,
1256 |         "book":"chapter_recommender-systems/autorec.html",
1257 |         "slides":["part-0.pdf",0],
1258 |         "slides_video":"",
1259 |         "notebook_video":"",
1260 |         "qa_video":""
1261 |     },
1262 |     {
1263 |         "title":"Personalized Ranking for Recommender Systems",
1264 |         "day_break":false,
1265 |         "book":"chapter_recommender-systems/ranking.html",
1266 |         "slides":["part-0.pdf",0],
1267 |         "slides_video":"",
1268 |         "notebook_video":"",
1269 |         "qa_video":""
1270 |     },
1271 |     {
1272 |         "title":"Neural Collaborative Filtering for Personalized Ranking",
1273 |         "day_break":false,
1274 |         "book":"chapter_recommender-systems/neumf.html",
1275 |         "slides":["part-0.pdf",0],
1276 |         "slides_video":"",
1277 |         "notebook_video":"",
1278 |         "qa_video":""
1279 |     },
1280 |     {
1281 |         "title":"Sequence-Aware Recommender Systems",
1282 |         "day_break":false,
1283 |         "book":"chapter_recommender-systems/seqrec.html",
1284 |         "slides":["part-0.pdf",0],
1285 |         "slides_video":"",
1286 |         "notebook_video":"",
1287 |         "qa_video":""
1288 |     },
1289 |     {
1290 |         "title":"Feature-Rich Recommender Systems",
1291 |         "day_break":false,
1292 |         "book":"chapter_recommender-systems/ctr.html",
1293 |         "slides":["part-0.pdf",0],
1294 |         "slides_video":"",
1295 |         "notebook_video":"",
1296 |         "qa_video":""
1297 |     },
1298 |     {
1299 |         "title":"Factorization Machines",
1300 |         "day_break":false,
1301 |         "book":"chapter_recommender-systems/fm.html",
1302 |         "slides":["part-0.pdf",0],
1303 |         "slides_video":"",
1304 |         "notebook_video":"",
1305 |         "qa_video":""
1306 |     },
1307 |     {
1308 |         "title":"Deep Factorization Machines",
1309 |         "day_break":false,
1310 |         "book":"chapter_recommender-systems/deepfm.html",
1311 |         "slides":["part-0.pdf",0],
1312 |         "slides_video":"",
1313 |         "notebook_video":"",
1314 |         "qa_video":""
1315 |     },
1316 |     {
1317 |         "title":"Generative Adversarial Networks",
1318 |         "day_break":false,
1319 |         "book":"chapter_generative-adversarial-networks/index.html",
1320 |         "slides":["part-0.pdf",0],
1321 |         "slides_video":"",
1322 |         "notebook_video":"",
1323 |         "qa_video":""
1324 |     },
1325 |     {
1326 |         "title":"Generative Adversarial Networks",
1327 |         "day_break":false,
1328 |         "book":"chapter_generative-adversarial-networks/gan.html",
1329 |         "slides":["part-0.pdf",0],
1330 |         "slides_video":"",
1331 |         "notebook_video":"",
1332 |         "qa_video":""
1333 |     },
1334 |     {
1335 |         "title":"Deep Convolutional Generative Adversarial Networks",
1336 |         "day_break":false,
1337 |         "book":"chapter_generative-adversarial-networks/dcgan.html",
1338 |         "slides":["part-0.pdf",0],
1339 |         "slides_video":"",
1340 |         "notebook_video":"",
1341 |         "qa_video":""
1342 |     },
1343 |     {
1344 |         "title":"Appendix: Mathematics for Deep Learning",
1345 |         "day_break":false,
1346 |         "book":"chapter_appendix-mathematics-for-deep-learning/index.html",
1347 |         "slides":["part-0.pdf",0],
1348 |         "slides_video":"",
1349 |         "notebook_video":"",
1350 |         "qa_video":""
1351 |     },
1352 |     {
1353 |         "title":"Geometry and Linear Algebraic Operations",
1354 |         "day_break":false,
1355 |         "book":"chapter_appendix-mathematics-for-deep-learning/geometry-linear-algebraic-ops.html",
1356 |         "slides":["part-0.pdf",0],
1357 |         "slides_video":"",
1358 |         "notebook_video":"",
1359 |         "qa_video":""
1360 |     },
1361 |     {
1362 |         "title":"Eigendecompositions",
1363 |         "day_break":false,
1364 |         "book":"chapter_appendix-mathematics-for-deep-learning/eigendecomposition.html",
1365 |         "slides":["part-0.pdf",0],
1366 |         "slides_video":"",
1367 |         "notebook_video":"",
1368 |         "qa_video":""
1369 |     },
1370 |     {
1371 |         "title":"Single Variable Calculus",
1372 |         "day_break":false,
1373 |         "book":"chapter_appendix-mathematics-for-deep-learning/single-variable-calculus.html",
1374 |         "slides":["part-0.pdf",0],
1375 |         "slides_video":"",
1376 |         "notebook_video":"",
1377 |         "qa_video":""
1378 |     },
1379 |     {
1380 |         "title":"Multivariable Calculus",
1381 |         "day_break":false,
1382 |         "book":"chapter_appendix-mathematics-for-deep-learning/multivariable-calculus.html",
1383 |         "slides":["part-0.pdf",0],
1384 |         "slides_video":"",
1385 |         "notebook_video":"",
1386 |         "qa_video":""
1387 |     },
1388 |     {
1389 |         "title":"Integral Calculus",
1390 |         "day_break":false,
1391 |         "book":"chapter_appendix-mathematics-for-deep-learning/integral-calculus.html",
1392 |         "slides":["part-0.pdf",0],
1393 |         "slides_video":"",
1394 |         "notebook_video":"",
1395 |         "qa_video":""
1396 |     },
1397 |     {
1398 |         "title":"Random Variables",
1399 |         "day_break":false,
1400 |         "book":"chapter_appendix-mathematics-for-deep-learning/random-variables.html",
1401 |         "slides":["part-0.pdf",0],
1402 |         "slides_video":"",
1403 |         "notebook_video":"",
1404 |         "qa_video":""
1405 |     },
1406 |     {
1407 |         "title":"Maximum Likelihood",
1408 |         "day_break":false,
1409 |         "book":"chapter_appendix-mathematics-for-deep-learning/maximum-likelihood.html",
1410 |         "slides":["part-0.pdf",0],
1411 |         "slides_video":"",
1412 |         "notebook_video":"",
1413 |         "qa_video":""
1414 |     },
1415 |     {
1416 |         "title":"Distributions",
1417 |         "day_break":false,
1418 |         "book":"chapter_appendix-mathematics-for-deep-learning/distributions.html",
1419 |         "slides":["part-0.pdf",0],
1420 |         "slides_video":"",
1421 |         "notebook_video":"",
1422 |         "qa_video":""
1423 |     },
1424 |     {
1425 |         "title":"Naive Bayes",
1426 |         "day_break":false,
1427 |         "book":"chapter_appendix-mathematics-for-deep-learning/naive-bayes.html",
1428 |         "slides":["part-0.pdf",0],
1429 |         "slides_video":"",
1430 |         "notebook_video":"",
1431 |         "qa_video":""
1432 |     },
1433 |     {
1434 |         "title":"Statistics",
1435 |         "day_break":false,
1436 |         "book":"chapter_appendix-mathematics-for-deep-learning/statistics.html",
1437 |         "slides":["part-0.pdf",0],
1438 |         "slides_video":"",
1439 |         "notebook_video":"",
1440 |         "qa_video":""
1441 |     },
1442 |     {
1443 |         "title":"Information Theory",
1444 |         "day_break":false,
1445 |         "book":"chapter_appendix-mathematics-for-deep-learning/information-theory.html",
1446 |         "slides":["part-0.pdf",0],
1447 |         "slides_video":"",
1448 |         "notebook_video":"",
1449 |         "qa_video":""
1450 |     },
1451 |     {
1452 |         "title":"Appendix: Tools for Deep Learning",
1453 |         "day_break":false,
1454 |         "book":"chapter_appendix-tools-for-deep-learning/index.html",
1455 |         "slides":["part-0.pdf",0],
1456 |         "slides_video":"",
1457 |         "notebook_video":"",
1458 |         "qa_video":""
1459 |     },
1460 |     {
1461 |         "title":"Using Jupyter",
1462 |         "day_break":false,
1463 |         "book":"chapter_appendix-tools-for-deep-learning/jupyter.html",
1464 |         "slides":["part-0.pdf",0],
1465 |         "slides_video":"",
1466 |         "notebook_video":"",
1467 |         "qa_video":""
1468 |     },
1469 |     {
1470 |         "title":"Using Amazon SageMaker",
1471 |         "day_break":false,
1472 |         "book":"chapter_appendix-tools-for-deep-learning/sagemaker.html",
1473 |         "slides":["part-0.pdf",0],
1474 |         "slides_video":"",
1475 |         "notebook_video":"",
1476 |         "qa_video":""
1477 |     },
1478 |     {
1479 |         "title":"Using AWS EC2 Instances",
1480 |         "day_break":false,
1481 |         "book":"chapter_appendix-tools-for-deep-learning/aws.html",
1482 |         "slides":["part-0.pdf",0],
1483 |         "slides_video":"",
1484 |         "notebook_video":"",
1485 |         "qa_video":""
1486 |     },
1487 |     {
1488 |         "title":"Using Google Colab",
1489 |         "day_break":false,
1490 |         "book":"chapter_appendix-tools-for-deep-learning/colab.html",
1491 |         "slides":["part-0.pdf",0],
1492 |         "slides_video":"",
1493 |         "notebook_video":"",
1494 |         "qa_video":""
1495 |     },
1496 |     {
1497 |         "title":"Selecting Servers and GPUs",
1498 |         "day_break":false,
1499 |         "book":"chapter_appendix-tools-for-deep-learning/selecting-servers-gpus.html",
1500 |         "slides":["part-0.pdf",0],
1501 |         "slides_video":"",
1502 |         "notebook_video":"",
1503 |         "qa_video":""
1504 |     },
1505 |     {
1506 |         "title":"Contributing to This Book",
1507 |         "day_break":false,
1508 |         "book":"chapter_appendix-tools-for-deep-learning/contributing.html",
1509 |         "slides":["part-0.pdf",0],
1510 |         "slides_video":"",
1511 |         "notebook_video":"",
1512 |         "qa_video":""
1513 |     },
1514 |     {
1515 |         "title":"Utility Functions and Classes",
1516 |         "day_break":false,
1517 |         "book":"chapter_appendix-tools-for-deep-learning/utils.html",
1518 |         "slides":["part-0.pdf",0],
1519 |         "slides_video":"",
1520 |         "notebook_video":"",
1521 |         "qa_video":""
1522 |     },
1523 |     {
1524 |         "title":"`d2l` API Document",
1525 |         "day_break":false,
1526 |         "book":"chapter_appendix-tools-for-deep-learning/d2l.html",
1527 |         "slides":["part-0.pdf",0],
1528 |         "slides_video":"",
1529 |         "notebook_video":"",
1530 |         "qa_video":""
1531 |     }
1532 | ]
1533 | 


--------------------------------------------------------------------------------
/favicon.ico:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/d2l-ai/courses-zh-v2/3d1b8b89eae0c836a24420051d33f2876a32e4d4/favicon.ico


--------------------------------------------------------------------------------
/generate.ipynb:
--------------------------------------------------------------------------------
  1 | {
  2 |  "cells": [
  3 |   {
  4 |    "cell_type": "markdown",
  5 |    "metadata": {},
  6 |    "source": [
  7 |     "Generate `contents.json` based on the `d2l-zh` repos, then manually copy and modify to `_modelules/*.json`"
  8 |    ]
  9 |   },
 10 |   {
 11 |    "cell_type": "code",
 12 |    "execution_count": null,
 13 |    "metadata": {},
 14 |    "outputs": [],
 15 |    "source": [
 16 |     "!pip install PyPDF2"
 17 |    ]
 18 |   },
 19 |   {
 20 |    "cell_type": "code",
 21 |    "execution_count": 1,
 22 |    "metadata": {
 23 |     "ExecuteTime": {
 24 |      "end_time": "2021-09-14T05:11:03.099469Z",
 25 |      "start_time": "2021-09-14T05:11:01.340659Z"
 26 |     }
 27 |    },
 28 |    "outputs": [],
 29 |    "source": [
 30 |     "import notedown\n",
 31 |     "import os\n",
 32 |     "import pathlib\n",
 33 |     "\n",
 34 |     "entry = '''    {\n",
 35 |     "        \"title\":\"TITLE\",\n",
 36 |     "        \"day_break\":false,\n",
 37 |     "        \"book\":\"URL\",\n",
 38 |     "        \"slides\":[\"part-0.pdf\",0],\n",
 39 |     "        \"slides_video\":\"\",\n",
 40 |     "        \"notebook_video\":\"\",\n",
 41 |     "        \"qa_video\":\"\"\n",
 42 |     "    }'''\n",
 43 |     "\n",
 44 |     "\n",
 45 |     "book_repo = pathlib.Path('/Users/mli/repos/d2l-en')\n",
 46 |     "\n",
 47 |     "def get_toc(root):\n",
 48 |     "    \"\"\"return a list of files in the order defined by TOC\"\"\"\n",
 49 |     "    subpages = _get_subpages(root)\n",
 50 |     "    res = [root]\n",
 51 |     "    for fn in subpages:\n",
 52 |     "        res.extend(get_toc(fn))\n",
 53 |     "    return res\n",
 54 |     "\n",
 55 |     "def _get_subpages(input_fn):\n",
 56 |     "    \"\"\"read toc in input_fn, returns what it contains\"\"\"\n",
 57 |     "    subpages = []\n",
 58 |     "    reader = notedown.MarkdownReader()\n",
 59 |     "    with open(input_fn, 'r', encoding='UTF-8') as f:\n",
 60 |     "        nb = reader.read(f)\n",
 61 |     "    for cell in nb.cells:\n",
 62 |     "        if (cell.cell_type == 'code' and 'attributes' in cell.metadata and\n",
 63 |     "                'toc' in cell.metadata.attributes['classes']):\n",
 64 |     "            for l in cell.source.split('\\n'):\n",
 65 |     "                l = l.strip()\n",
 66 |     "                if not l.startswith(':'):\n",
 67 |     "                    fn = os.path.join(os.path.dirname(input_fn), l + '.md')\n",
 68 |     "                    if os.path.exists(fn):\n",
 69 |     "                        subpages.append(fn)\n",
 70 |     "    return subpages\n",
 71 |     "\n",
 72 |     "def _get_title(filename):\n",
 73 |     "    with open(filename, 'r') as f:\n",
 74 |     "        lines = f.readlines()\n",
 75 |     "        for l in lines:\n",
 76 |     "            if l.startswith('#'): return l[1:].strip()\n",
 77 |     "\n",
 78 |     "entries = []\n",
 79 |     "notebooks = get_toc(str(book_repo/'index.md'))\n",
 80 |     "for nb in notebooks:\n",
 81 |     "    p = str(pathlib.Path(nb).relative_to(book_repo).with_suffix('.html'))\n",
 82 |     "    if 'index.md' in p: continue\n",
 83 |     "    title = _get_title(nb)\n",
 84 |     "    if not title: continue\n",
 85 |     "    entries.append(entry.replace('TITLE', title).replace('URL', p))\n",
 86 |     "\n",
 87 |     "with open('contents.json', 'w') as f:\n",
 88 |     "    f.write('[\\n' + ',\\n'.join(entries) + '\\n]\\n')\n"
 89 |    ]
 90 |   },
 91 |   {
 92 |    "cell_type": "code",
 93 |    "execution_count": 2,
 94 |    "metadata": {
 95 |     "ExecuteTime": {
 96 |      "end_time": "2021-09-14T05:11:03.116157Z",
 97 |      "start_time": "2021-09-14T05:11:03.101253Z"
 98 |     }
 99 |    },
100 |    "outputs": [],
101 |    "source": [
102 |     "from PyPDF2 import PdfFileReader, PdfFileWriter\n",
103 |     "import subprocess\n",
104 |     "\n",
105 |     "def extract_pdf(source, start_page, end_page, target):\n",
106 |     "    source = pathlib.Path(source)\n",
107 |     "    target = pathlib.Path(target)\n",
108 |     "    if target.exists() and target.stat().st_mtime > source.stat().st_mtime:\n",
109 |     "        return\n",
110 |     "    pdf = PdfFileReader(str(source))\n",
111 |     "    assert end_page > start_page\n",
112 |     "    assert end_page <= pdf.getNumPages()\n",
113 |     "    pdf_writer = PdfFileWriter()\n",
114 |     "    for page in range(start_page, end_page):\n",
115 |     "        pdf_writer.addPage(pdf.getPage(page))\n",
116 |     "    with open('/tmp/tmp.pdf', 'wb') as out:\n",
117 |     "        pdf_writer.write(out)\n",
118 |     "    # compress pdf size\n",
119 |     "    # refer to https://askubuntu.com/questions/113544/how-can-i-reduce-the-file-size-of-a-scanned-pdf-file\n",
120 |     "    cmd = f'gs -sDEVICE=pdfwrite -dCompatibilityLevel=1.4 -dPDFSETTINGS=/prepress -dNOPAUSE  -dBATCH -sOutputFile={str(target)} /tmp/tmp.pdf'\n",
121 |     "    process = subprocess.Popen(cmd, shell=True, stdout=subprocess.PIPE,\n",
122 |     "                               stderr=subprocess.PIPE)\n",
123 |     "    stdout, _ = process.communicate()\n",
124 |     "    if process.returncode != 0:\n",
125 |     "        print(stdout.decode().splitlines())\n",
126 |     "    print(f'Written {end_page-start_page} pages to {target}')"
127 |    ]
128 |   },
129 |   {
130 |    "cell_type": "markdown",
131 |    "metadata": {},
132 |    "source": [
133 |     "Generate all `_modules/part*.md`"
134 |    ]
135 |   },
136 |   {
137 |    "cell_type": "code",
138 |    "execution_count": 4,
139 |    "metadata": {
140 |     "ExecuteTime": {
141 |      "end_time": "2021-09-14T05:16:04.145250Z",
142 |      "start_time": "2021-09-14T05:16:03.932538Z"
143 |     },
144 |     "scrolled": true
145 |    },
146 |    "outputs": [],
147 |    "source": [
148 |     "import datetime\n",
149 |     "import pathlib\n",
150 |     "import json\n",
151 |     "import os\n",
152 |     "\n",
153 |     "\n",
154 |     "pdf_dir = '/Users/mli/Google Drive/d2l-zh-v2-slides/'\n",
155 |     "slides_dir = 'assets/pdfs/'\n",
156 |     "notebooks_dir = 'assets/notebooks/'\n",
157 |     "book_url = 'https://zh-v2.d2l.ai/'\n",
158 |     "notebook_url = 'https://nbviewer.jupyter.org/format/slides/github/d2l-ai/d2l-zh-pytorch-slides/blob/main/'\n",
159 |     "notbook_repo = '../d2l-zh-pytorch-slides/'\n",
160 |     "video_url = 'https://www.bilibili.com/video/'\n",
161 |     "cur_day = datetime.datetime(2021, 3, 19)\n",
162 |     "\n",
163 |     "titles = ['深度学习基础', '卷积神经网络', '计算机视觉', '循环神经网络', '注意力机制',]\n",
164 |     "holidays = [datetime.datetime(2021, m, d) for m, d in ((4,3),(4,4),(4,25),(5,1),(5,2),(5,8),(5,9),(6,12),(6,13),(7,31),(8,1))]\n",
165 |     "slide_pages = {}\n",
166 |     "\n",
167 |     "for i, title in enumerate(titles):\n",
168 |     "    p = pathlib.Path('_modules')\n",
169 |     "    with (p/f'part_{i}.md').open('w') as f:\n",
170 |     "        f.write(f'---\\ntitle: {title}\\n---\\n')\n",
171 |     "        if not (p/f'part_{i}.json').exists():\n",
172 |     "            contents = []\n",
173 |     "        else:\n",
174 |     "            contents = json.load((p/f'part_{i}.json').open('r'))\n",
175 |     "        for entry in contents:\n",
176 |     "            if entry['day_break']:\n",
177 |     "                while True:\n",
178 |     "                    if cur_day.weekday() == 6: # sun\n",
179 |     "                        cur_day += datetime.timedelta(days=6)\n",
180 |     "                    else:\n",
181 |     "                        cur_day += datetime.timedelta(days=1)\n",
182 |     "                    f.write(f'\\n{cur_day.month}月{cur_day.day}日\\n\\n')\n",
183 |     "                    if cur_day not in holidays:\n",
184 |     "                        break\n",
185 |     "                    f.write(': **休课**{: .label .label-green }\\n')\n",
186 |     "            # title\n",
187 |     "            f.write(f': {entry[\"title\"]}\\n')\n",
188 |     "            # html page\n",
189 |     "            if 'book' in entry and entry['book']:\n",
190 |     "                f.write(f'  : [<span class=\"iconfont icon-xiaoshuo-copy\"></span>]({book_url+entry[\"book\"]})\\n')\n",
191 |     "            else:\n",
192 |     "                f.write('  : &nbsp; \\n')\n",
193 |     "            # pdf\n",
194 |     "            pdf, page = entry['slides']\n",
195 |     "            if page:\n",
196 |     "                if pdf not in slide_pages:\n",
197 |     "                    slide_pages[pdf] = [0,]\n",
198 |     "                save_pdf = f'{slides_dir}part-{i}_{len(slide_pages[pdf])}.pdf'\n",
199 |     "                cur_page = sum(slide_pages[pdf])\n",
200 |     "                extract_pdf(pdf_dir+pdf, cur_page, cur_page+page, save_pdf)\n",
201 |     "                slide_pages[pdf].append(page)\n",
202 |     "                f.write(f'  : [<span class=\"iconfont icon-KeynoteOutline\"></span>]({save_pdf})\\n')\n",
203 |     "            else:\n",
204 |     "                f.write('  : &nbsp; \\n')\n",
205 |     "\n",
206 |     "            # notebook\n",
207 |     "            write_notebook = False\n",
208 |     "            if 'book' in entry and entry['book'] and  (not 'notebook' in  entry or entry['notebook']):\n",
209 |     "                notebook_path = entry[\"book\"].replace('.html', '.ipynb')\n",
210 |     "                notebook_file = notbook_repo + notebook_path\n",
211 |     "                notebook_output = pathlib.Path(notebooks_dir + notebook_path).with_suffix('.slides.html')\n",
212 |     "                if os.path.exists(notebook_file):\n",
213 |     "                    if not notebook_output.exists():\n",
214 |     "                        os.system(f'jupyter nbconvert {notebook_file} --to slides --output-dir {str(notebook_output.parent)}')\n",
215 |     "                    if notebook_output.exists():\n",
216 |     "                        write_notebook = True\n",
217 |     "                        f.write(f'  : [<span class=\"iconfont icon-jupyter\"></span>]({str(notebook_output)})\\n')\n",
218 |     "            if not write_notebook:\n",
219 |     "                f.write('  :  &nbsp; \\n')\n",
220 |     "                    #if entry['notebook_video']:\n",
221 |     "                    #    f.write(f' [<span style=\"font-size:130%\"  class=\"iconfont icon-bilibili-fill\"></span> 代码]({entry[\"notebook_video\"]}) &nbsp;')\n",
222 |     "            #if 'qa_video' in entry and entry['qa_video']:\n",
223 |     "            #   f.write(f' [<span style=\"font-size:130%\"  class=\"iconfont icon-bilibili-fill\"></span> 问答]({entry[\"qa_video\"]}) &nbsp;')\n",
224 |     "            if entry['slides_video']:\n",
225 |     "                f.write(f'  : [<span style=\"font-size:130%\"  class=\"iconfont icon-bilibili-fill\"></span>]({entry[\"slides_video\"]})\\n')\n",
226 |     "            else:\n",
227 |     "                f.write('  :  &nbsp; \\n')\n",
228 |     "            f.write('\\n')\n",
229 |     "\n",
230 |     "\n",
231 |     "!touch index.md\n"
232 |    ]
233 |   },
234 |   {
235 |    "cell_type": "code",
236 |    "execution_count": 2,
237 |    "metadata": {
238 |     "ExecuteTime": {
239 |      "end_time": "2021-07-16T05:57:54.653894Z",
240 |      "start_time": "2021-07-16T05:57:54.648149Z"
241 |     }
242 |    },
243 |    "outputs": [
244 |     {
245 |      "name": "stdout",
246 |      "output_type": "stream",
247 |      "text": [
248 |       "Overwriting video\n"
249 |      ]
250 |     }
251 |    ],
252 |    "source": [
253 |     "%%writefile video\n",
254 |     "6\t488\n",
255 |     "14\t813\n",
256 |     "2\t966\n",
257 |     "5\t281\n",
258 |     "17\t875\t1\n",
259 |     "5\t446\t1\n",
260 |     "14\t543\n",
261 |     "19\t771\t1\n",
262 |     "10\t758\n",
263 |     "15\t699\n",
264 |     "7\t402\t1\n",
265 |     "11\t897\n",
266 |     "6\t503\n",
267 |     "8\t902\t1\n",
268 |     "6\t402\t1\n",
269 |     "11\t626\n",
270 |     "7\t384\n",
271 |     "6\t564\t1\n",
272 |     "16\t1123\t1\n",
273 |     "4\t249\t1\n",
274 |     "11\t832\n",
275 |     "13\t1359\n",
276 |     "7\t505\t1\n",
277 |     "7\t1116\n",
278 |     "10\t1006\n",
279 |     "8\t407\t1\n",
280 |     "6\t781\n",
281 |     "8\t754\t1\n",
282 |     "6\t719\n",
283 |     "6\t736\t1\n",
284 |     "10\t902\n",
285 |     "12\t1443\n",
286 |     "13\t870\t1\n",
287 |     "2\t519\n",
288 |     "6\t807\t1\n",
289 |     "10\t1039\t1\n",
290 |     "4\t328\t1\n",
291 |     "5\t363\t1\n",
292 |     "8\t814\t1\n",
293 |     "5\t998\n",
294 |     "9\t661\n",
295 |     "7\t899\n",
296 |     "8\t607\t1\n",
297 |     "8\t847\n",
298 |     "3\t304\t1\n",
299 |     "10\t1172\n",
300 |     "3\t395\t1\n",
301 |     "7\t620\n",
302 |     "4\t417\n",
303 |     "6\t510\n",
304 |     "6\t1306\t1\n",
305 |     "13\t2115\n",
306 |     "4\t638\t1\n",
307 |     "7\t543\n",
308 |     "5\t442\t1\n",
309 |     "6\t654\n",
310 |     "4\t600\t1\n",
311 |     "15\t1618\n",
312 |     "4\t429\t1\n",
313 |     "6\t1078\n",
314 |     "7\t1031\t1\n",
315 |     "9\t849\n",
316 |     "6\t564\t1\n",
317 |     "2\t396\n",
318 |     "14\t2361\n",
319 |     "16\t1758\n",
320 |     "5\t1300\n",
321 |     "5\t721\t1\n",
322 |     "16\t1304\n",
323 |     "9\t1238\n",
324 |     "14\t768\t1\n",
325 |     "9\t832\n",
326 |     "8\t534\t1\n",
327 |     "13\t1676\t1\n",
328 |     "10\t1148\t1\n",
329 |     "6\t695\n",
330 |     "4\t227\t1\n",
331 |     "6\t585\t1\n",
332 |     "7\t1086\n",
333 |     "18\t1400\t1\n",
334 |     "16\t2524\n",
335 |     "5\t596\t1\n",
336 |     "14\t3165\t1\n",
337 |     "5\t431\n",
338 |     "10\t1676\t1\n",
339 |     "3\t560\n",
340 |     "5\t760\t1\n",
341 |     "2\t262\n",
342 |     "11\t1060\t1\n",
343 |     "3\t297\n",
344 |     "8\t1157\t1"
345 |    ]
346 |   },
347 |   {
348 |    "cell_type": "code",
349 |    "execution_count": 10,
350 |    "metadata": {
351 |     "ExecuteTime": {
352 |      "end_time": "2021-07-16T06:01:44.518853Z",
353 |      "start_time": "2021-07-16T06:01:44.495121Z"
354 |     }
355 |    },
356 |    "outputs": [
357 |     {
358 |      "name": "stdout",
359 |      "output_type": "stream",
360 |      "text": [
361 |       "total: 48\n",
362 |       "avg page: 8.395833333333334\n",
363 |       "avg length: 15.159722222222223\n",
364 |       "page per min: 2.045208144478978\n",
365 |       "total: 43\n",
366 |       "avg page: 8.0\n",
367 |       "avg length: 13.234883720930233\n",
368 |       "page per min: 1.7313444384536163\n"
369 |      ]
370 |     }
371 |    ],
372 |    "source": [
373 |     "with open('video') as f:\n",
374 |     "    lines = f.readlines()\n",
375 |     "    tokens = [[int(i) for i in l.strip().split('\\t')] for l in lines]\n",
376 |     "nbs = [x for x in tokens if len(x)==2]\n",
377 |     "code = [x[:2] for x in tokens if len(x)==3]\n",
378 |     "\n",
379 |     "import numpy as np\n",
380 |     "\n",
381 |     "def analy(x):\n",
382 |     "    x = np.array(x)\n",
383 |     "    print('total:', len(x))\n",
384 |     "    print('avg page:', x[:,0].mean())\n",
385 |     "    print('avg length:', x[:,1].mean()/60)\n",
386 |     "    print('page per min:', (x[:,1]/x[:,0]).mean()/60)\n",
387 |     "\n",
388 |     "analy(nbs)\n",
389 |     "analy(code)"
390 |    ]
391 |   },
392 |   {
393 |    "cell_type": "code",
394 |    "execution_count": null,
395 |    "metadata": {},
396 |    "outputs": [],
397 |    "source": []
398 |   }
399 |  ],
400 |  "metadata": {
401 |   "kernelspec": {
402 |    "display_name": "Python 3",
403 |    "language": "python",
404 |    "name": "python3"
405 |   },
406 |   "language_info": {
407 |    "codemirror_mode": {
408 |     "name": "ipython",
409 |     "version": 3
410 |    },
411 |    "file_extension": ".py",
412 |    "mimetype": "text/x-python",
413 |    "name": "python",
414 |    "nbconvert_exporter": "python",
415 |    "pygments_lexer": "ipython3",
416 |    "version": "3.8.5"
417 |   }
418 |  },
419 |  "nbformat": 4,
420 |  "nbformat_minor": 5
421 | }
422 | 


--------------------------------------------------------------------------------
/index.md:
--------------------------------------------------------------------------------
 1 | ---
 2 | layout: home
 3 | title: 课程安排
 4 | ---
 5 | 
 6 | # 动手学深度学习在线课程
 7 | 
 8 | ## 课程更新
 9 | 
10 | {% if site.announcements %}
11 | {{ site.announcements.last }}
12 | [所有公告](announcements.html){: .btn .fs-3 }
13 | {% endif %}
14 | 
15 | <!-- <div class="responsive-video-container">
16 | 
17 | <iframe src="//player.bilibili.com/player.html?aid=289532467&bvid=BV1if4y147hS&cid=309491732&page=1" scrolling="no" border="0" frameborder="no" framespacing="0" allowfullscreen="true"> </iframe>
18 | 
19 | </div>  -->
20 | 
21 | ## 课程摘要
22 | 
23 | | --- | --- |
24 | | 课时  | 2021年3月20日 --- （预计）7月 |
25 | | 直播时间 | 北京时间每周六、日下午 1:00 --- 2:30 |
26 | | 直播地址 | [<span style="font-size:140%" class="iconfont icon-livestreaming"></span> 机器之心](https://app6ca5octe2206.h5.xiaoeknow.com/v1/course/alive/l_601cc496e4b05a9e88714463) |
27 | | 视频回放 | [<span style="font-size:140%" class="iconfont icon-bilibili-fill"></span> B站](https://space.bilibili.com/1567748478) |
28 | | 教材 | [<span class="iconfont icon-xiaoshuo-copy"></span> zh-v2.d2l.ai](https://zh-v2.d2l.ai/) |
29 | 
30 | 
31 | 
32 | 不论是在学术突破还是在工业应用，
33 | 深度学习是人工智能在近十年里进展最为迅速的领域。然而，深度学习模型复杂、参数繁多、而且新模型层出不穷，这给学习带来了难度。
34 | 
35 | 本课程将从零开始教授深度学习。同学们只需要有基础的Python编程和数学基础。我们将覆盖四大类模型：多层感知机、卷积神经网络、循环神经网络、和注意力机制。在此之上，我们将介绍深度学习中的两大应用领域---计算机视觉和自然语言处理---中的典型任务。
36 | 
37 | 本课程的一大特点是不仅讲述模型算法，同时会将每一处细节都讲述如何用PyTorch进行实现。这样同学们可以在真实数据上获得第一手经验。我们将举办四次课程竞赛，让同学们实践学习到的知识如何解决实际问题。
38 | 
39 | 课程内容将紧靠《动手学深度学习》第二版。本书目前已经被近200所大学采用作为教材。我们将在3月20日开始直播。同学们无需注册或缴费就可以参加。敬请期待。
40 | 
41 | 
42 | 
43 | 
44 | ## 讲师
45 | 
46 | {% for staffer in site.staffers %}
47 | {{ staffer }}
48 | {% endfor %}
49 | 
50 | <div style="clear: both;"></div>
51 | 
52 | 
53 | 
54 | 
55 | ## 课程安排
56 | 
57 | 
58 | 目前日程安排为暂定安排，会根据实际进度进行调整。部分内容暂时链接到英文版，中文版会随后更新。
59 | 
60 | 
61 | {% for module in site.modules %}
62 | {{ module }}
63 | {% endfor %}
64 | 


--------------------------------------------------------------------------------
/upload.sh:
--------------------------------------------------------------------------------
1 | aws s3 sync --delete _site s3://courses.d2l.ai/zh-v2 --acl 'public-read'
2 | 


--------------------------------------------------------------------------------