-
Notifications
You must be signed in to change notification settings - Fork 0
/
search.xml
234 lines (112 loc) · 75.9 KB
/
search.xml
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
<?xml version="1.0" encoding="utf-8"?>
<search>
<entry>
<title>2022南燕研途指引未来线上求职分享会</title>
<link href="/posts/a627.html"/>
<url>/posts/a627.html</url>
<content type="html"><![CDATA[<h3 id="0-背景"><a href="#0-背景" class="headerlink" title="0 背景"></a>0 背景</h3><p>本次求职分享会由北京大学深研院和共青团北大深研院举办,涉及体制内外两大专场,主体涵盖体制、编制、金融、互联网等不同行业,<a href="https://mp.weixin.qq.com/s/7LEPCMsctIyXqA6snqxEFA">官方微信推送及各位嘉宾介绍此链接</a>。</p><div><!--块级封装--> <center><!--将图片和文字居中--> <img src="https://cdn.jsdelivr.net/gh/sqzhang-jeremy/CDN/post_images/2021/202204040951973.jpeg" alt="无法显示图片时显示的文字" style="zoom:80%"> <br><!--换行--> 图1. 求职分享会主题海报 [1]<!--标题--> </center></div><h3 id="1-“考公选择与成公交流”"><a href="#1-“考公选择与成公交流”" class="headerlink" title="1 “考公选择与成公交流”"></a>1 “考公选择与成公交流”</h3><p>作为一名曾经律师从业人员和税务系统从业人员,赵学长首先在开场对体制内的岗位分布和履行职能进行介绍,跟我们描绘了一幅体制内的图景;近些年全国各地招生和实习生工作经历分享;以及结合体制内工作八年的经历分享一些工作经验和求职建议。</p><h4 id="1-1什么是公务员?"><a href="#1-1什么是公务员?" class="headerlink" title="1.1什么是公务员?"></a>1.1什么是公务员?</h4><div><!--块级封装--> <center><!--将图片和文字居中--> <img src="https://cdn.jsdelivr.net/gh/sqzhang-jeremy/CDN/post_images/2021/202204041451625.png" alt="无法显示图片时显示的文字" style="zoom:40%"> <br><!--换行--> 图1. 赵学长对公务员的定义<!--标题--> </center></div><p>公务员(Civil servants),全称为国家公务员,是各国负责统筹管理<u>经济社会秩序和国家公共资源,维护国家法律规定和贯彻执行相关义务</u>的公职人员。在中国,公务员是指依法履行公职、纳入国家行政编制和由国家财政负担工资福利的工作人员。</p><div><!--块级封装--> <center><!--将图片和文字居中--> <img src="https://cdn.jsdelivr.net/gh/sqzhang-jeremy/CDN/post_images/2021/202204041505704.png" alt="无法显示图片时显示的文字" style="zoom:33%"> <br><!--换行--> 图3. 中华人民共和国行政区划(2018年) [2]<!--标题--> </center></div><p>要想了解体制内的工作分布和履行职能,首先要了解我国的政治体制和各职能部门。我国政府管理体系包括五级:中央-省-市-县区-乡镇,这一管理体系最早可追溯于秦统一六国形成的郡县制三级管理体系:中央-省-郡县。按照这一地理层级中央与地方的划分,再加上我国是由中国共产党绝对政府的绝对领导,我国形成了以国家为核心的党政、人大监察/法院/监察机构的权力机构,省/市/县直接复制国家的管理机构,乡镇有所调整。依照现有治理体系,分别开设相应职能部门。根据岗位和职能不同,体制内岗位分布大体包括:国家公务员考试(国考)、省市级公务员考试(省/市考)、选调生考试、部队文职、事业单位考试等。</p><div><!--块级封装--> <center><!--将图片和文字居中--> <img src="https://cdn.jsdelivr.net/gh/sqzhang-jeremy/CDN/post_images/2021/202204041048210.jpg" alt="无法显示图片时显示的文字" style="zoom:70%"> <br><!--换行--> 图4. 我国治理体系 [3]<!--标题--> </center></div><h3 id="参考"><a href="#参考" class="headerlink" title="参考"></a>参考</h3><p>[1] <a href="https://cdn.jsdelivr.net/gh/sqzhang-jeremy/CDN/post_images/2021/202204040951973.jpeg">求职分享会主题海报</a></p><p>[2] 兰小欢,《置身事内:中国政府与经济发展》,上海人民出版社,2021</p><p>[3] <a href="https://www.zhihu.com/question/33538093/answer/634814066">知乎用户@老夫子在“国考与省考,二者的待遇,发展空间等有什么区别?</a></p><p>[0] cover image:<a href="https://www.bitgab.com/uploads/1580271660-Job-Search-Cartoon-1580271660.jpg">https://www.bitgab.com/uploads/1580271660-Job-Search-Cartoon-1580271660.jpg</a></p>]]></content>
<categories>
<category> Lecture </category>
</categories>
<tags>
<tag> Job </tag>
</tags>
</entry>
<entry>
<title>2021中外高校毕业生职业发展研究与展望--领英中国&全球化智库</title>
<link href="/posts/7e5c.html"/>
<url>/posts/7e5c.html</url>
<content type="html"><![CDATA[<h1 id="《高校校友观察:中外高校毕业生职业发展研究与展望-2021》及个人所思"><a href="#《高校校友观察:中外高校毕业生职业发展研究与展望-2021》及个人所思" class="headerlink" title="《高校校友观察:中外高校毕业生职业发展研究与展望 2021》及个人所思"></a>《高校校友观察:中外高校毕业生职业发展研究与展望 2021》及个人所思</h1><h2 id="1背景"><a href="#1背景" class="headerlink" title="1背景"></a>1背景</h2><blockquote><p>新冠肺炎疫情的全球大流行使世界经济遭遇历史性衰退,劳动力市场也面临严峻挑战。中国作为世界第二大经济体,在疫情防控与经济复苏中率先取得阶段性成效,经济的稳步恢复与发展拉动了就业市场的持续回暖。但是,由于就业岗位供给情况仍受到外部环境复杂多变和未来市场不确定性的影响,2021年全国普通高校毕业生规模又达到<strong>909</strong>万人的历史新高,在此背景下,我国就业市场的供需两端均面临较大压力。</p><p>此次是全球化智库(CCG)与领英中国(LinkedIn)《高校校友观察》系列报告的第二期,在第一期聚焦于中国U10 毕业生职业发展的基础上,<u>将研究范围扩大至全球100所高校,对比了中外高校毕业生在深造、就业及国际迁移等方面的异同,深入探讨了中外顶尖高校人才职业发展的特征及其影响因素,为了解新时期国际人才流动趋势、调整和制定人才发展战略提供了有益参考</u>。</p></blockquote><p><img src="https://cdn.jsdelivr.net/gh/sqzhang-jeremy/CDN/post_images/2021/202111201615548.png"></p><h3 id="1-1研究目标"><a href="#1-1研究目标" class="headerlink" title="1.1研究目标"></a>1.1研究目标</h3><p>此报告希望能为以下群体和目标提供参考:</p><ul><li>高校毕业生的职业规划</li><li>高等教育机构的发展与人才培养</li><li>政府出台相关政策</li></ul><h2 id="2报告"><a href="#2报告" class="headerlink" title="2报告"></a>2报告</h2><h3 id="2-1目录"><a href="#2-1目录" class="headerlink" title="2.1目录"></a>2.1目录</h3><p><img src="https://cdn.jsdelivr.net/gh/sqzhang-jeremy/CDN/post_images/2021/202111201044129.png"></p><h3 id="2-2概况"><a href="#2-2概况" class="headerlink" title="2.2概况"></a>2.2概况</h3><p>G100: 2022年QS排名前100 (LinkedIn用户,样本量约250w)</p><p>U10: top10国内高校(样本包括所有在LinkedIn填写U10教育背景,并且在14-19年间毕业用户,样本量约为19w)</p><p><img src="https://cdn.jsdelivr.net/gh/sqzhang-jeremy/CDN/post_images/2021/202111201616292.png"></p><p><img src="https://cdn.jsdelivr.net/gh/sqzhang-jeremy/CDN/post_images/2021/202111201049593.png"></p><p><img src="https://cdn.jsdelivr.net/gh/sqzhang-jeremy/CDN/post_images/2021/202111201049670.png"></p><p><img src="https://cdn.jsdelivr.net/gh/sqzhang-jeremy/CDN/post_images/2021/202111201050083.png"></p><h3 id="2-3深造"><a href="#2-3深造" class="headerlink" title="2.3深造"></a>2.3深造</h3><p><img src="https://cdn.jsdelivr.net/gh/sqzhang-jeremy/CDN/post_images/2021/202111201051086.png"></p><p><img src="https://cdn.jsdelivr.net/gh/sqzhang-jeremy/CDN/post_images/2021/202111201052299.png"></p><p><img src="https://cdn.jsdelivr.net/gh/sqzhang-jeremy/CDN/post_images/2021/202111201053410.png"></p><p>U10更多的是去发达地区进修,G100更多留在本国(本地高校排名和质量)</p><p><img src="https://cdn.jsdelivr.net/gh/sqzhang-jeremy/CDN/post_images/2021/202111201055189.png"></p><p><img src="https://cdn.jsdelivr.net/gh/sqzhang-jeremy/CDN/post_images/2021/202111201058273.png"></p><p><img src="https://cdn.jsdelivr.net/gh/sqzhang-jeremy/CDN/post_images/2021/202111201101084.png"></p><h3 id="2-4职场"><a href="#2-4职场" class="headerlink" title="2.4职场"></a>2.4职场</h3><p><img src="https://cdn.jsdelivr.net/gh/sqzhang-jeremy/CDN/post_images/2021/202111201102939.png"></p><p><img src="https://cdn.jsdelivr.net/gh/sqzhang-jeremy/CDN/post_images/2021/202111201106106.png"></p><p><img src="https://cdn.jsdelivr.net/gh/sqzhang-jeremy/CDN/post_images/2021/202111201106290.png"></p><p><img src="https://cdn.jsdelivr.net/gh/sqzhang-jeremy/CDN/post_images/2021/202111201817110.png"></p><p><img src="https://cdn.jsdelivr.net/gh/sqzhang-jeremy/CDN/post_images/2021/202111201111317.png"></p><p>发达国家对发展中国家吸引力很大</p><p><img src="https://cdn.jsdelivr.net/gh/sqzhang-jeremy/CDN/post_images/2021/202111201113486.png"> </p><p><img src="https://cdn.jsdelivr.net/gh/sqzhang-jeremy/CDN/post_images/2021/202111201114525.png"></p><p><img src="https://cdn.jsdelivr.net/gh/sqzhang-jeremy/CDN/post_images/2021/202111201115919.png"></p><p>针对这一项,其实今年中国在社区和社会服务上有很多漂亮的工作,像战胜疫情、脱贫等,都是很好的社区和社会服务的案例,这一块儿还需要在国际化方面需要进一步加强</p><p><img src="https://cdn.jsdelivr.net/gh/sqzhang-jeremy/CDN/post_images/2021/202111201123872.png"></p><p>![image-20211120112415205](C:/Users/zhang/Library/Application Support/typora-user-images/image-20211120112415205.png)</p><p><img src="https://cdn.jsdelivr.net/gh/sqzhang-jeremy/CDN/post_images/2021/202111201124451.png"></p><h3 id="2-5国际迁移"><a href="#2-5国际迁移" class="headerlink" title="2.5国际迁移"></a>2.5国际迁移</h3><p><img src="https://cdn.jsdelivr.net/gh/sqzhang-jeremy/CDN/post_images/2021/202111201128296.png"></p><p> <img src="https://cdn.jsdelivr.net/gh/sqzhang-jeremy/CDN/post_images/2021/202111201130534.png"></p><p>国内经济向好</p><p><img src="https://cdn.jsdelivr.net/gh/sqzhang-jeremy/CDN/post_images/2021/202111201132478.png"></p><p>新时代多次迁徙,毕业->回国,毕业->当地就业->回国</p><p><img src="https://cdn.jsdelivr.net/gh/sqzhang-jeremy/CDN/post_images/2021/202111201137338.png"></p><p>就业行业很接近</p><h3 id="2-6建议"><a href="#2-6建议" class="headerlink" title="2.6建议"></a>2.6建议</h3><p><img src="https://cdn.jsdelivr.net/gh/sqzhang-jeremy/CDN/post_images/2021/202111201140301.png"></p><p><img src="https://cdn.jsdelivr.net/gh/sqzhang-jeremy/CDN/post_images/2021/202111201143769.png"></p><h2 id="3讨论"><a href="#3讨论" class="headerlink" title="3讨论"></a>3讨论</h2><p>Q围绕着以下三点:</p><ul><li>1如何看待国际毕业生和国内毕业生在深造、就业和国际迁移方面的异同?</li><li>2中外顶尖高校对职业发展的影响因素有哪些?</li><li>3政策/毕业生职业发展等方面,提出建议?</li></ul><p>A我提炼了一些讨论的结论:</p><ul><li><p>1数据可依赖性</p><ul><li>国内外分别约19w和250w的数据量</li><li>从大样本考虑,复合统计学规律,选择服从正态分布</li></ul></li><li><p>2行业选择</p><ul><li><p>2.1国内外高度重合,前三热门工作职能:<u>工程、研究和教育</u></p></li><li><p>2.2国内外行业分布:</p><ul><li>国内<ul><li>互联网/信息科技行业</li><li>金融服务行业</li><li>受疫情影响的医疗相关行业</li></ul></li><li>国外<ul><li>教育</li><li>国际金融服务行业:四大</li><li>互联网/信息科技行业</li></ul></li></ul></li></ul></li><li><p>3就业方面思考</p><ul><li>3.1信息焦虑:信息太多,选择太多</li><li>3.2个人发展<ul><li>1时间管理</li><li>2个人性格+职业热情</li><li>3长短期目标</li><li>4科学的建议:职业教练</li></ul></li></ul></li><li><p>4FIRE</p><p>FIRE (Financial Independence, Retire Early 经济独立,提前退休) 运动 是一种以经济独立(又叫财务自由)和提前退休为目标,以及重视幸福感多于对物质的满足的一种生活方式。</p><ul><li>‘我想1年过365天,不想1天过365次’</li></ul></li><li><p>5竞争力</p><ul><li>5.1人多。中国人多,无论到哪儿都竞争,希望通过深造提高竞争力</li><li>5.2认识世界。完全在国内学习,很难对整个世界进行认知,而通过深造的跨境流动是一种很好的选择</li><li>5.3中国人的惯性。一方面是优秀学生的惯性:本硕博,周围的影响和带动,另一方面是报送机制</li></ul></li></ul><h2 id="4个人思考"><a href="#4个人思考" class="headerlink" title="4个人思考"></a>4个人思考</h2><p>做决策时的选择,要考虑方方面面:</p><ul><li>1自我方面<ul><li>希望自己和各位能在接受这么多年的应试教育后,开始关注更多<u>自我方面</u>的思考:你是什么样的性格?你的兴趣是什么?你为什么选择当前要从事的行业?是否从事的行业和你的兴趣相契合?你会感到快乐么?</li><li>以下是关于MBTI性格测试的链接:<a href="https://www.16personalities.com/free-personality-test">https://www.16personalities.com/free-personality-test</a> (我的性格测试结果是:ENFJ-A)</li></ul></li><li>2调研<ul><li>专业角度<ul><li>权威期刊论文</li><li>大学网站相关课程介绍</li><li>维基百科/知乎等</li><li>Youtube/bibilibi/播客/自媒体博主</li></ul></li><li>多维度:<ul><li>从事该行业的人员,像刚就业/本硕博在读/已经毕业多年的人</li><li>动用自己的人脉资源,打听打听从事该行业的相关熟悉/陌生的人</li><li>其他跟该行业有相关度的人员,像接触过这个领域的人/参加过实验/游戏参与</li></ul></li></ul></li><li>3LinkedIn<ul><li>了解该专业毕业生的过往经历、项目经历、技能需求和就业分布</li><li>一方面是养成习惯丰富职业过往和不断提升自己的技能,另一方面是增加潜在就业机会</li><li>我的LinkedIn: <a href="https://www.linkedin.com/in/shiquan-zhang-b97815222/">https://www.linkedin.com/in/shiquan-zhang-b97815222/</a>, 欢迎connect</li></ul></li></ul><h2 id="5参考"><a href="#5参考" class="headerlink" title="5参考"></a>5参考</h2><p>[1]《高校校友观察:中外高校毕业生职业发展研究与展望 2021》官方直播:<a href="https://k.cnki.net/CLectureLive/Index/16443">https://k.cnki.net/CLectureLive/Index/16443</a></p><p>[2] 封面:<a href="https://www.topresume.com/career-advice/management-career-path">https://www.topresume.com/career-advice/management-career-path</a></p><p>[3] FIRE1: <a href="https://zh.wikipedia.org/wiki/FIRE%E8%BF%90%E5%8A%A8">https://zh.wikipedia.org/wiki/FIRE%E8%BF%90%E5%8A%A8</a></p><p>[4] FIRE2: <a href="https://zhuanlan.zhihu.com/p/217775503">https://zhuanlan.zhihu.com/p/217775503</a></p><p>[5] Personality test: <a href="https://www.16personalities.com/free-personality-test">https://www.16personalities.com/free-personality-test</a></p>]]></content>
<categories>
<category> Lecture </category>
</categories>
<tags>
<tag> Talk </tag>
</tags>
</entry>
<entry>
<title>Ubuntu 18.04 install</title>
<link href="/posts/5ee5.html"/>
<url>/posts/5ee5.html</url>
<content type="html"><![CDATA[<h1 id="Ubuntu-18-04-install"><a href="#Ubuntu-18-04-install" class="headerlink" title="Ubuntu 18.04 install"></a>Ubuntu 18.04 install</h1><h2 id="0-Check电脑内置引导启动"><a href="#0-Check电脑内置引导启动" class="headerlink" title="0 Check电脑内置引导启动"></a>0 Check电脑内置引导启动</h2><p>[检查电脑内置引导启动]这一步骤很重要,它关系到<u>后续的电脑启动引导模式</u>和<u>Ubuntu安装界面</u>。</p><pre><code># Windows1 Win+R2 type msinfo323 check BIOS mode</code></pre><p><img src="https://cdn.jsdelivr.net/gh/sqzhang-jeremy/CDN/post_images/2021/202109231046503.png"></p><p>开机引导启动有常用的两种:Legacy, UEFI(像我的电脑还支持双模式)</p><p>Legacy BIOS:慢慢就要淘汰了,主要是慢</p><p>UEFI:统一可扩展固件接口(Unified Extensible Firmware Interface, UEFI)是一种将计算机固件连接到其操作系统的软件程序规范; 比较简单的理解是快</p><p>具体的内容可以参考知乎:<a href="https://www.zhihu.com/question/21672895">UEFI 引导与 BIOS 引导在原理上有什么区别?</a></p><h2 id="1-下载镜像-制作启动盘-分配空间"><a href="#1-下载镜像-制作启动盘-分配空间" class="headerlink" title="1 下载镜像+制作启动盘+分配空间"></a>1 下载镜像+制作启动盘+分配空间</h2><h3 id="1-1-下载连接"><a href="#1-1-下载连接" class="headerlink" title="1.1 下载连接"></a>1.1 下载连接</h3><p>Ubuntu18.04LTS: <a href="https://releases.ubuntu.com/18.04/">https://releases.ubuntu.com/18.04/</a></p><p>UltraIOS: <a href="https://cn.ultraiso.net/xiazai.html">https://cn.ultraiso.net/xiazai.html</a></p><h3 id="1-2-制作启动盘"><a href="#1-2-制作启动盘" class="headerlink" title="1.2 制作启动盘"></a>1.2 制作启动盘</h3><p>具体制作启动盘参考ref1, 里面很详细</p><h3 id="1-3-磁盘分区"><a href="#1-3-磁盘分区" class="headerlink" title="1.3 磁盘分区"></a>1.3 磁盘分区</h3><p> 在分配磁盘空间时,要耐心等待(电脑根据自身硬件不同,速度存在差异)</p><h2 id="2-Windows引导或者手动安装分区"><a href="#2-Windows引导或者手动安装分区" class="headerlink" title="2 Windows引导或者手动安装分区"></a>2 Windows引导或者手动安装分区</h2><h3 id="2-1-Windows引导"><a href="#2-1-Windows引导" class="headerlink" title="2.1 Windows引导"></a>2.1 Windows引导</h3><p><img src="https://cdn.jsdelivr.net/gh/sqzhang-jeremy/CDN/post_images/2021/202109231047637.jpg"></p><p><img src="https://cdn.jsdelivr.net/gh/sqzhang-jeremy/CDN/post_images/2021/202109231047585.jpg"></p><p>在这里,如果你提前在<u>1.3磁盘分区</u>中提前预留好一部分未分配的空间,此时选择第一个 install Ubuntu alongside Windows Boot manager时,系统会默认自动分配下面系统所需要的:/, Swap, /boot, /home的空间,同时系统在开机的时候会变成Ubuntu引导;否则选择第五个,手动分区,见下面<u>2.2手动安装分区</u></p><h4 id="2-1-1-Legacy-BIOS-system"><a href="#2-1-1-Legacy-BIOS-system" class="headerlink" title="2.1.1 Legacy BIOS system"></a>2.1.1 Legacy BIOS system</h4><p><img src="https://cdn.jsdelivr.net/gh/sqzhang-jeremy/CDN/post_images/2021/202109231048718.jpg"></p><h4 id="2-1-2-UEFI"><a href="#2-1-2-UEFI" class="headerlink" title="2.1.2 UEFI"></a>2.1.2 UEFI</h4><p><img src="https://cdn.jsdelivr.net/gh/sqzhang-jeremy/CDN/post_images/2021/202109231048513.jpg"></p><p>不要问为啥,我俩都知道….我一开始对这个概念不重视,后面就翻车了…..<strong>所以这里再次强调一开始check电脑内置引导启动的方式很重要</strong></p><h3 id="2-2-手动安装分区"><a href="#2-2-手动安装分区" class="headerlink" title="2.2 手动安装分区"></a>2.2 手动安装分区</h3><p>Ubuntu磁盘分区(预算500G的配置)</p><table><thead><tr><th align="center">分区名称</th><th align="center">描述</th><th align="center">大小</th><th align="center">分区设置分区选择</th><th align="center">分区选择</th></tr></thead><tbody><tr><td align="center">/</td><td align="center">C盘:安装系统和软件</td><td align="center">100G</td><td align="center">主分区</td><td align="center">/</td></tr><tr><td align="center">Swap</td><td align="center">虚拟内存</td><td align="center">16G</td><td align="center">逻辑分用</td><td align="center">用于交换</td></tr><tr><td align="center">/boot</td><td align="center">linux内核及引导程序</td><td align="center">1G</td><td align="center">主分区</td><td align="center">/boot</td></tr><tr><td align="center">/home</td><td align="center">用户工作目录</td><td align="center">300G</td><td align="center">逻辑分区</td><td align="center">/home</td></tr><tr><td align="center">/temp</td><td align="center">缓存</td><td align="center">4G</td><td align="center">逻辑分区</td><td align="center">/temp</td></tr></tbody></table><p>留100G左右保留暂不使用,以备其他后续用途</p><h2 id="3-解决开机停留在-ok-紫色屏幕页面"><a href="#3-解决开机停留在-ok-紫色屏幕页面" class="headerlink" title="3 解决开机停留在[ok]紫色屏幕页面"></a>3 解决开机停留在[ok]紫色屏幕页面</h2><p>原因:Ubuntu内置显卡驱动与电脑显卡驱动不匹配,导致无法进入图像界面</p><h3 id="3-1-方案一"><a href="#3-1-方案一" class="headerlink" title="3.1 方案一"></a>3.1 方案一</h3><p>解决方案思路:进入Ubuntu命令行-卸载Ubuntu内置显卡驱动-进入Ubuntu低分辨率页面-安装新的显卡驱动</p><p>核心步骤:进入Ubuntu加载页面,按[‘e’]进入编辑模式,在倒数第二行后添加[‘nomodeset’],再按ctrl+x进入命令模式</p><pre><code>1在命令行界面登录用户2卸载所有旧驱动程序sudo apt-get remove nvidia-*sudo apt-get autoremove3通过.run文件卸载驱动sudo nvidia-uninstall4重启系统,正常进入图形界面,但由于此时没有显卡驱动,系统分辨率较低。sudo reboot</code></pre><h3 id="3-2-方案二"><a href="#3-2-方案二" class="headerlink" title="3.2 方案二"></a>3.2 方案二</h3><p>解决方案思路:在启动页面,进入Ubuntu高级模式下的安全模式(recovery mode),安装适配电脑系统的Nivida显卡</p><p>上述两种方案解决的都是:装机结束后,当前系统的显卡驱动软件的版本和配置,与现有的实际显卡驱动不一致造成的问题。(推荐方案二)</p><p>实在解决不了,可以看这个小哥的视频:<a href="https://www.youtube.com/watch?v=Tg4fWsFPzSE">https://www.youtube.com/watch?v=Tg4fWsFPzSE</a></p><h2 id="4-参考"><a href="#4-参考" class="headerlink" title="4 参考"></a>4 参考</h2><p>[0]<a href="https://www.itzgeek.com/how-tos/linux/ubuntu-how-tos/how-to-install-ubuntu-18-04-alongside-with-windows-10-or-8-in-dual-boot.html">https://www.itzgeek.com/how-tos/linux/ubuntu-how-tos/how-to-install-ubuntu-18-04-alongside-with-windows-10-or-8-in-dual-boot.html</a></p><p>[1]<a href="https://blog.csdn.net/baidu_36602427/article/details/86548203">https://blog.csdn.net/baidu_36602427/article/details/86548203</a></p><p>[2]<a href="https://www.youtube.com/watch?v=Tg4fWsFPzSE">https://www.youtube.com/watch?v=Tg4fWsFPzSE</a></p><p>[3]<a href="https://www.jb51.net/article/187465.htm">https://www.jb51.net/article/187465.htm</a></p><p>[4]<a href="https://whatis.techtarget.com/definition/Unified-Extensible-Firmware-Interface-UEFI#:~:text=Unified%20Extensible%20Firmware%20Interface%20(UEFI)%20is%20a%20specification%20for%20a,its%20operating%20system%20(OS).&text=Like%20BIOS%2C%20UEFI%20is%20installed,runs%20when%20booting%20a%20computer">https://whatis.techtarget.com/definition/Unified-Extensible-Firmware-Interface-UEFI#:~:text=Unified%20Extensible%20Firmware%20Interface%20(UEFI)%20is%20a%20specification%20for%20a,its%20operating%20system%20(OS).&text=Like%20BIOS%2C%20UEFI%20is%20installed,runs%20when%20booting%20a%20computer</a></p><p>Cover image:<a href="https://images.app.goo.gl/UhQQHsacUG2x8dEy6">https://images.app.goo.gl/UhQQHsacUG2x8dEy6</a></p>]]></content>
<categories>
<category> Software </category>
</categories>
<tags>
<tag> Ubuntu </tag>
</tags>
</entry>
<entry>
<title>博客描述</title>
<link href="/posts/9b15.html"/>
<url>/posts/9b15.html</url>
<content type="html"><![CDATA[<h2 id="1前言"><a href="#1前言" class="headerlink" title="1前言"></a>1前言</h2><p>本博客通过hexo搭建站点,利用PicGo将图片同步至GitHub repo作为CDN加速,然后部署在GitHub上,博客主题:<a href="https://github.com/blinkfox/hexo-theme-matery">matery</a></p><p>搭建一个博客真的很耗费时间(改前端代码UI)但是会很有成就感,同时也是一种督促自己学习记录的一个地方。刚开始我参考网上主流的搭建博客方法:hexo+github部署,随后由于访问速度问题,切换到aliyun/coding/码云上,最后才部署到服务器上。网站部署到github上尤其要注意一点:网站必须为**xxxx.github.io(**xxxx一定要为你的repository的名字,否则就gg了,我就是一个活生生的反例,搞了半个小时才恍然大悟)</p><pre><code># Update process# 清空之前的文件/缓存1hexo clean # 根据配置文件渲染出一套静态页面2hexo g# 渲染局部动态网站html3hexo s # 将上一步渲染出的一系列文件上传至github4hexo d / hexo g -d</code></pre><h2 id="2博客功能更新时间线"><a href="#2博客功能更新时间线" class="headerlink" title="2博客功能更新时间线"></a>2博客功能更新时间线</h2><p>待更新…….</p><h2 id="3-参考"><a href="#3-参考" class="headerlink" title="3 参考"></a>3 参考</h2><p>[1]<a href="https://yafine-blog.cn/posts/4ab2.html">https://yafine-blog.cn/posts/4ab2.html</a></p>]]></content>
<categories>
<category> blog </category>
</categories>
<tags>
<tag> hexo </tag>
</tags>
</entry>
<entry>
<title>我的第一篇博文</title>
<link href="/posts/6fdc.html"/>
<url>/posts/6fdc.html</url>
<content type="html"><![CDATA[<h1 id="20200308-第一篇博文"><a href="#20200308-第一篇博文" class="headerlink" title="20200308 第一篇博文"></a>20200308 第一篇博文</h1><p>终于把大体的框架搭好了,要开始写博客了,该走的弯路一点都没少……</p><p>为啥我老是那么喜欢较真,认准的事情一定要做下去,太难了☹️☹️☹️</p> <li> hexo clean # 1Update process 清空之前的文件/缓存 </li> <li> hexo g # 2根据配置文件渲染出一套静态页面 </li> <li> hexo s # 3渲染局部动态网站html </li> <li> 4hexo d //hexo g -d # 4将上一步渲染出的一系列文件上传至github </li> <h1 id="20230308-更新部署网站"><a href="#20230308-更新部署网站" class="headerlink" title="20230308 更新部署网站"></a>20230308 更新部署网站</h1><p>针对删除网站原始文件后,遇到如下问题:</p><pre class="line-numbers language-bash"><code class="language-bash">kex_exchange_identification: Connection closed by remote host Connection closed by 198.18.0.166 port 22fatal: Could not <span class="token function">read</span> from remote repository.Please <span class="token function">make</span> sure you have the correct access rights and the repository exists.<span aria-hidden="true" class="line-numbers-rows"><span></span><span></span><span></span><span></span></span></code></pre><p>如何做来解决</p><pre class="line-numbers language-bash"><code class="language-bash"><span class="token comment" spellcheck="true"># 1 install hexo-deployer-git</span><span class="token function">npm</span> <span class="token function">install</span> hexo-deployer-git --save 安装 hexo-deployer-git 此步骤只需要做一次<span class="token punctuation">[</span>1<span class="token punctuation">]</span><span class="token comment" spellcheck="true"># 2 ask help from New Bing</span>Hello, this is Bing. I'm sorry to hear that you encountered a bug when trying to deploy your code to GitHub.😞There could be several possible causes <span class="token keyword">for</span> this error, such as:- Your SSH <span class="token function">service</span> is not operational or listening on port 22¹².- You have incorrect access rights or repository name¹.- You are using the wrong IP address <span class="token punctuation">(</span>server IP instead of private IP<span class="token punctuation">)</span> to connect³.To fix this bug, you may want to try some of the following steps:- Check your SSH <span class="token function">service</span> status and port number with <span class="token variable"><span class="token variable">`</span><span class="token function">ps</span> aux <span class="token operator">|</span> <span class="token function">grep</span> <span class="token function">ssh</span><span class="token variable">`</span></span> and <span class="token variable"><span class="token variable">`</span><span class="token function">netstat</span> -plant <span class="token operator">|</span> <span class="token function">grep</span> :22<span class="token variable">`</span></span> commands².- Verify your SSH key and repository URL with <span class="token variable"><span class="token variable">`</span><span class="token function">ssh</span> -T [email protected]<span class="token variable">`</span></span> and <span class="token variable"><span class="token variable">`</span><span class="token function">git</span> remote -v<span class="token variable">`</span></span> commands¹.- Use your private IP instead of server IP <span class="token keyword">if</span> applicable³.I hope this helps. Please <span class="token keyword">let</span> me know <span class="token keyword">if</span> you need <span class="token function">more</span> assistance.😊Source: Conversation with Bing, 3/8/2023<span class="token punctuation">(</span>1<span class="token punctuation">)</span> kex_exchange_identification: Connection closed by remote host - GIT AND <span class="token punctuation">..</span><span class="token punctuation">..</span> https://stackoverflow.com/questions/74444666/kex-exchange-identification-connection-closed-by-remote-host-git-and-github Accessed 3/8/2023.<span class="token punctuation">(</span>2<span class="token punctuation">)</span> ssh_exchange_identification: Connection closed by remote host under Git <span class="token punctuation">..</span><span class="token punctuation">..</span> https://stackoverflow.com/questions/10127818/ssh-exchange-identification-connection-closed-by-remote-host-under-git-bash Accessed 3/8/2023.<span class="token punctuation">(</span>3<span class="token punctuation">)</span> What causes SSH error: kex_exchange_identification: Connection closed <span class="token punctuation">..</span><span class="token punctuation">..</span> https://serverfault.com/questions/1015547/what-causes-ssh-error-kex-exchange-identification-connection-closed-by-remote Accessed 3/8/2023.<span aria-hidden="true" class="line-numbers-rows"><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span></span></code></pre><h1 id="参考"><a href="#参考" class="headerlink" title="参考"></a>参考</h1><p>[1] <a href="https://ryanluoxu.github.io/2017/11/24/%E7%94%A8-Hexo-%E5%92%8C-GitHub-Pages-%E6%90%AD%E5%BB%BA%E5%8D%9A%E5%AE%A2/">将本地 Hexo 博客部署在 Github 上</a></p>]]></content>
<categories>
<category> blog </category>
</categories>
<tags>
<tag> hexo </tag>
</tags>
</entry>
<entry>
<title>现代医学和公共卫生对人类的影响</title>
<link href="/posts/8057.html"/>
<url>/posts/8057.html</url>
<content type="html"><![CDATA[<h1 id="现代医学的诞生和大众卫生的前世今生"><a href="#现代医学的诞生和大众卫生的前世今生" class="headerlink" title="现代医学的诞生和大众卫生的前世今生"></a>现代医学的诞生和大众卫生的前世今生</h1><p><img src="https://cdn.jsdelivr.net/gh/sqzhang-jeremy/CDN/post_photosIMG_0064.jpg"></p><p>昨天在微博上看到男神<strong>吴军</strong>老师有新的直播,第一时间搬好小板凳静静地涨知识。今天整理笔记、谈谈自己的看法和顺便练习Markdown语法。</p><p>此次直播主题是:现代医学的诞生和公共卫生的前世今生。主要内容包括以下三点以及答疑环节。</p><ul><li>①现代医学的前世今生及其科学性</li><li>②大众卫生对国家和人民健康的影响</li><li>③大众卫生学科和医学有什么区别和联系</li></ul><h2 id="1-现代医学的前世今生"><a href="#1-现代医学的前世今生" class="headerlink" title="1 现代医学的前世今生"></a>1 现代医学的前世今生</h2><h3 id="1-1-定义"><a href="#1-1-定义" class="headerlink" title="1.1 定义"></a>1.1 定义</h3><p>在现代医学诞生之前, 人们一直依赖着那种「时灵时不灵」的方法而生存。古有劳民伤财炼制长生不老丹,今有挥斥巨资寄托人体冷冻术。中医具有悠久的历史,在中华民族五千年的文明中遍布足迹。随着西医慢慢映入眼帘,人们更倾向于理性科学地对面对问题,中医的可解释性备受质疑。但这并不等于扼杀了中医在整个医学史上的地位。在我看来,中医感性,西医理性。中医和西医是相辅相成的,各有优劣,唯有中西结合方可延年益寿。</p><blockquote><p>医学,《辞海》的解释是:研究人类生命过程及防治疾病的科学体系。从人的整体性及其同外界环境的辨证关系出发,用实验研究,现场调查,临床观察等方法,研究人类生命活动和外界环境的互相关系,人类疾病的发生,发展及其防治规律,以及增进健康,延长寿命和提高劳动能力的有效措施。我国医学有<strong>传统医学</strong>和<strong>现代医学</strong>两个体系。《中国大百科全书》也把医学分为<strong>中国传统医学</strong>和<strong>现代医学</strong>两个部分。</p></blockquote><h3 id="1-2-历史"><a href="#1-2-历史" class="headerlink" title="1.2 历史"></a>1.2 历史</h3><p>约2500年前,古希腊医学家希波克拉底提出了<strong>四体液学说</strong>(人体四体液紊乱导致疾病),强调疾病不是外邪入侵,而是人体自身紊乱。</p><p>约1800年前,古罗马帝国医学家盖伦继承并进一步丰富了四体液学说。在动物解剖实验的基础上,提出<strong>三精气学说、血液耗散与循环学说</strong>,迫使西方医疗停滞近1500年,导致<strong>放血疗法</strong>大行其道。(著名钢琴家莫扎特、美国首任总统华盛顿等均遭受放血致死)</p><p>约500年前,比利时医学家维萨里通过解剖人体深入了解人体结构,著书<strong>《人体构造》</strong>并进一步提出对盖伦学说的质疑。</p><p>约400年前,英国医学家哈维划通过解剖提出了<strong>血液循环学说</strong>、通过实验和逻辑推理著书<strong>《心血运动论》</strong>,完全否定了盖伦的理论。他划时代地创举为现代医学对人体生理与疾病的认识进一步奠定基础。此时慢慢出现了理性之光,对医学来说,以自然为师而不是哲学为师。同时,利用逻辑推理启发了著名科学家笛卡尔。</p><blockquote><p><strong>笛卡尔的科学方法</strong></p><ul><li>提出问题: 比解决问题还重要</li><li>进行实验: 发展到以后的双盲对比实验</li><li>结论和解释: 光得到结论还不行,还要解释,具有支撑的解释才能形成理论</li><li>推广结论并普遍化</li><li>在实践中发现新的问题,如此往复循环</li></ul></blockquote><p>通过实验+逻辑推理的这套方法论,我们沿用至今。医学的进步,是可以复制的。通过科学的方法论,加上经验的累积,还有运气偶然因素,进而导致结果的必然。</p><p>哈维之后的历史这里就没有深究,从那时至今,又是一段很长很长的历史。从细菌的发现、现代外科手术、消毒、青霉素、抗生素、DNA、青蒿素、听诊器、医学影像(US、CT/PET、MRI)以及现在炙手可热的大数据技术……</p><p><img src="https://cdn.jsdelivr.net/gh/sqzhang-jeremy/CDN/post_photoswuyingdengxia.png"></p><p>如果大家感兴趣,欢迎去了解。推荐一部去年国产的纪录片<a href="https://www.bilibili.com/bangumi/media/md28222504">《无影灯下/手术两百年/200 YEARS OF SURGERY》</a>。评分高达9.4,看完后我很震撼,感叹文明和理性之光,同时由衷地表达对医学和医生的尊重和敬意,希望以后能为医学领域的发展贡献自己的一份力量。</p><p>P.S.: 对MRI成像作出杰出贡献的Paul C.Lauterbur在2003年获得诺贝尔生理或医学奖,Mr.Lauterbur将该奖章捐赠予SIAT(SIAT-Paul C.Lauterbur Biomedical Imaging Center)激励中国发展医学影像技术。</p><h3 id="1-3-对医学的态度"><a href="#1-3-对医学的态度" class="headerlink" title="1.3 对医学的态度"></a>1.3 对医学的态度</h3><p>我们生活在一个伟大的时代,时代也因我们的存在而伟大。人类对于医学的态度是随着科学技术的发展而不断变化的。尤其是近百年来仪器的发展与进步,使得我们可以借助外物看清楚身体内的变化,这为医学的诊断提供了有力的支撑和参考。短短的几十年间,人们对于健康的态度已发生转变:从持续几千年的治疗、救死扶伤到现在以预防为主,使得人类的寿命得以延续。温饱、膳食均衡、讲卫生、定期体检、注射疫苗等,时至今日,看病已经是我们面对生死的最后一步。</p><p>现在,科学研究发现影响人体寿命的因素,不仅仅是器官的衰退老化,还有生活环境、饮食、作息、心态…… 随着科技的进步与发展,很多技术已经改变了我们的生活。新的医学仪器会继续产生,在这个智能时代,24h实施对人体进行监控,通过对人体海量数据的分析来预防疾病,未来某天像特斯拉汽车每一个零部件都装有传感器实时反馈信息的案例定会发生在我们人类身上。在未来,我相信我们会更加感性和理性地对待医学和生命。</p><p>从第一声啼哭到平静离去,一睁眼一闭眼的功夫,一辈子就过去了。诞生是奇迹,而离去又让我们无能为力。敬畏生命🙏🙏🙏</p><h2 id="2公共卫生"><a href="#2公共卫生" class="headerlink" title="2公共卫生"></a>2公共卫生</h2><h3 id="2-1-定义"><a href="#2-1-定义" class="headerlink" title="2.1 定义"></a>2.1 定义</h3><p>吴军老师是这样定义的:</p><blockquote><p>公共卫生是促进和保护人类的生命以及他们生活、学习、工作、娱乐等的健康。</p></blockquote><p> <a href="https://www.jhsph.edu/about/what-is-public-health/">Johns Hopkins Bloomberg School of Public Health</a>(全美公共卫生专业第一)是这样定义的:</p><blockquote><p><strong>Protecting Health, Saving Lives—<em>Millions at a Time</em></strong> </p><p>In the medical field, clinicians treat diseases and injuries one patient at a time. <strong>But in public health, we <em>prevent</em> disease and injury</strong>. Public health researchers, practitioners and educators work with <em>communities</em> and <em>populations</em>. <strong>We identify the <em>causes</em> of disease and disability, and we implement largescale solutions.</strong> Today, public health encompasses areas as wideranging as epigenetics, chronic disease, the science of aging, mental health, disaster response, refugee health, injury prevention and tobacco control.</p></blockquote><p>可见,公共卫生是大众卫生,担当健康促进和疾病预防的使命。它不是医学专业,而是从整个卫生和医学角度去考虑人类的健康问题。面对这次新型冠状病毒疫情,应对措施中 我们不仅涉及医学,还涉及政府的管理协调、全国人民的积极配合等。我们没有前瞻性和应急措施,暴露出我国在类似公共卫生领域发展的不足、人才的匮乏。因此,近期才出台相关政策鼓励类似领域、人才的发展和培养。</p><h3 id="2-2-历史"><a href="#2-2-历史" class="headerlink" title="2.2 历史"></a>2.2 历史</h3><p>公共卫生的历史相比现代医学那可短多了,将近300年。</p><p>在17-18世纪,前身。受瘟疫、霍乱、天花等流行病的影响,公众渐渐引起重视。在当时应对流行病,国家政府执行隔离、检疫措施;美国东部城市设立流行病委员会;推广疫苗。</p><p>在19世纪,伟大的卫生觉醒。英国泰晤士河开始进行清理,清理环境和减轻污染;监督供水和废物处理;对细菌的了解提高全社会的卫生水平;明确社会责任,生病不再是一个人的事情。</p><p>在20世纪,社会生活的一部分。护理行业的出现、社会福利(针对母婴)、对精神病患者的的救助和照顾、公共卫生常识的教育、公共卫生预算大幅度提高、习惯的养成(受西班牙流感影响,人们随地吐痰的现象得到遏制)</p><p>为了面对疫情、减轻疫情、控制疫情,在这个过程中产生了公共卫生。目前公共卫生在西方国家普及率比较高,也越来越引起人们的重视,但是像我就是第一次听说🤐🤐🤐</p><h3 id="2-3-从业者"><a href="#2-3-从业者" class="headerlink" title="2.3 从业者"></a>2.3 从业者</h3><p>急救人员(消防人员)、餐厅检察员(预防好多疾病)、科学家和研究人员、营养学家(营养学和PH结合)、公共卫生医生和护士、流行病学家、健康教育者(学校讲生理卫生的老师)、社区规划师、职业健康与安全人员、公共政策制定者、卫生师(第三国家,赤脚医生)</p><h3 id="2-4-发展的好处"><a href="#2-4-发展的好处" class="headerlink" title="2.4 发展的好处"></a>2.4 发展的好处</h3><p>虽然着科技的进步和人们生活水平的逐渐提高,大众卫生和未来医疗逐渐引起人们的重视。在如今的医疗条件下,人们的平均寿命相比几十年前,已经得到了较大的延长。想必大家都想活得健康、活得开心、活得长久?只有当公共卫生得到大面积的普及推广、突发应急措施的完善、全民重视和身体力行,我们才能减少疫情大面积爆发,更好地生存活着,更好地面对未来。</p><h2 id="3-中国未来展望"><a href="#3-中国未来展望" class="headerlink" title="3 中国未来展望"></a>3 中国未来展望</h2><ul><li>1政府在公共卫生领域有很大的投入</li><li>2全社会的卫生意识会得到加强,公共卫生成为国民素质教育的一部分</li><li>3很多相关产业的出现(防护产品、远程教育办公等)</li></ul><h2 id="4-Q-amp-A"><a href="#4-Q-amp-A" class="headerlink" title="4 Q&A"></a>4 Q&A</h2><p>由于Q&A环节涉及到太多问题了,我挑选几个代表性的问题。</p><p>Q1: 大数据AI将会对现有的医疗带来什么变革?取代医生?</p><p>A1: 不会,医生并不多,反而不够用。大数据会推动医疗的进步。之前在《智能时代》书中也有提到过(这是我当时在未来医疗章节做的笔记):</p><p><img src="https://cdn.jsdelivr.net/gh/sqzhang-jeremy/CDN/post_photos20200308174538.png"></p><p>Q2: 传统医学的副作用?是药三分毒?</p><p>A2: 像有些药品用了几十年后才发现致癌。接触医生比较多,医生建议还是少吃药,尽可能利用自身免疫系统。</p><p>Q3: 像这样大规模的流行病,如何发现首例0号,如何应对?</p><p>A3: 很难说首例是怎么发现的。像西班牙流感、黑死病,到今天为止仍未解,找到源头蛮困难的。</p><p>Q4: 超级传播者如何组织和预防?</p><p>A4: 在病毒的进化过程中,毒性较强的往往把宿主搞死,传播距离不会太远,病毒的寿命也会比较短;相反,毒性较弱的,传播比较远,不会造成太大的伤害。</p><p>Q5: 大数据如何帮助减轻危重病人的死亡率</p><p>A5: 举一个例子,微软受博爱公司委托目前的一个项目。进入ICU,防止进入ICU/第二次进入;通过大数据经验,及时采取措施,降低再次进入的概率。</p><p>Q6: 人工智能在医学领域的进展?未来发展</p><p>A6: </p><ul><li>1手术机器人。医生在做手术时容易抖动,即使是再有经验的医生,在手术视频放大的情况下,仍然能观测到细微的抖动镜头。目前手术机器人的精确度相当高,眼科手术机器人精度可以达到千分之二。推荐大家了解一下JHU的达芬奇手术机器人。</li><li>2疑难杂症的检测。之所是疑难杂症,是因为再有经验的医生,见得少。</li><li>3 临时突发公共卫生事件,医疗资源不够,AI通过大数据提供建议。</li></ul><p>Q7: 如何能够提高的全名公共卫生意识?</p><p>A7: </p><ul><li>1政府慢慢设立专门的部分</li><li>2慢慢素质教育增加</li><li>3公益直播,公共卫生的普及教育</li></ul><p>Q8: 您是如何评价远程办公和远程教学?未来领域常态化?</p><p>A8: 后续会有文章谈论远程办公、远程教育。</p><ul><li>1远程教学一直是手段,今年是远程教学的元年(就像2003年是阿里的电商元年)。</li><li>2在线教育,可以大数据实时反馈信息。</li><li>3远程办公效率不高。绝大部分,在家办公,无非是时间福利,效率不高,项目最关键的一段时间是大家一起在房间里(Warroom)。之前我担任谷歌搜索部高官时,同时管四个不同地区明显感觉不太好处理。例子:雅虎,员工根本没登陆到公司,短期可以,长期不行。管理者可以采取一些措施:早请示,晚汇报;仪式感;早晨工作安排汇报。</li></ul><p>再次放一张男神吴军老师的照片。</p><p><img src="https://cdn.jsdelivr.net/gh/sqzhang-jeremy/CDN/post_photosIMG_0069.jpg"></p><p>本片文章来自于吴军老师在高山大学直播平台的线上直播、本人看法并加以整理。</p><p>References:</p><p>[1]<a href="https://www.sohu.com/a/255961875_799846">https://www.sohu.com/a/255961875_799846</a> </p><p>[2]<a href="http://www.360doc.com/content/18/0722/19/13975006_772422274.shtml">http://www.360doc.com/content/18/0722/19/13975006_772422274.shtml</a></p><p>[3]<a href="http://blog.sina.com.cn/s/blog_672a43c70102vaqz.html">http://blog.sina.com.cn/s/blog_672a43c70102vaqz.html</a></p><p>[4]<a href="https://www.bilibili.com/bangumi/play/ep284560">https://www.bilibili.com/bangumi/play/ep284560</a></p><p>[5]<a href="https://www.jhsph.edu/about/what-is-public-health/">https://www.jhsph.edu/about/what-is-public-health/</a></p><p>Cover image: <a href="https://pulitzercenter.org/campus-consortium/johns-hopkins-bloomberg-school-public-health">https://pulitzercenter.org/campus-consortium/johns-hopkins-bloomberg-school-public-health</a></p>]]></content>
<categories>
<category> Lecture </category>
</categories>
<tags>
<tag> Talk </tag>
</tags>
</entry>
<entry>
<title>Pytorch 数据加载器源码分析</title>
<link href="/posts/2004.html"/>
<url>/posts/2004.html</url>
<content type="html"><![CDATA[<p>最近在优化程序的代码,于是借此机会加深对框架底层代码的认识。</p><h2 id="1DataLoader介绍"><a href="#1DataLoader介绍" class="headerlink" title="1DataLoader介绍"></a>1DataLoader介绍</h2><h3 id="1-1定义"><a href="#1-1定义" class="headerlink" title="1.1定义"></a>1.1定义</h3><p>官方解释:“数据加载由数据集和采样器组成,基于python的单、多进程的iterators来处理数据。”</p><p><code>DataLoader()</code>是Pytorch中数据读取的一个重要接口。该接口在<code>dataloader.py</code>文件中,通常用Pytorch来训练模型基本都会用到,像tensorflow、mxnet等框架类似。该接口的目的:将自定义的dataset根据batch size、是否shuffle等封装成一个batch sieze的Tensor用于训练模型。通俗地讲:数据喂给模型的接口,我们设置好如何喂、喂多少等,批量封装成一个用于训练的特有东西。</p><h3 id="1-2-数据加载流程"><a href="#1-2-数据加载流程" class="headerlink" title="1.2 数据加载流程"></a>1.2 数据加载流程</h3><ul><li>创建一个Dataset对象</li><li>创建一个DataLoader对象</li><li>循环这个DataLoader对象, 将img, lable加载到模型中训练</li></ul><pre class="line-numbers language-python"><code class="language-python">dataset <span class="token operator">=</span> MyDataset<span class="token punctuation">(</span><span class="token punctuation">)</span>dataloader <span class="token operator">=</span> DataLoader<span class="token punctuation">(</span>dataset<span class="token punctuation">)</span><span class="token keyword">for</span> epoch <span class="token keyword">in</span> range<span class="token punctuation">(</span>num_epochs<span class="token punctuation">)</span><span class="token punctuation">:</span> <span class="token keyword">for</span> img<span class="token punctuation">,</span> label <span class="token keyword">in</span> dataloader<span class="token punctuation">:</span> <span class="token punctuation">.</span><span class="token punctuation">.</span><span class="token punctuation">.</span><span class="token punctuation">.</span><span class="token punctuation">.</span><span class="token punctuation">.</span><span aria-hidden="true" class="line-numbers-rows"><span></span><span></span><span></span><span></span><span></span></span></code></pre><h2 id="2源码"><a href="#2源码" class="headerlink" title="2源码"></a>2源码</h2><pre class="line-numbers language-python"><code class="language-python"><span class="token keyword">class</span> <span class="token class-name">DataLoader</span><span class="token punctuation">(</span>object<span class="token punctuation">)</span><span class="token punctuation">:</span> r<span class="token triple-quoted-string string">""" Data loader. Combines a dataset and a sampler, and provides single- or multi-process iterators over the dataset. Arguments: dataset (Dataset): dataset from which to load the data. batch_size (int, optional): how many samples per batch to load (default: 1). shuffle (bool, optional): set to ``True`` to have the data reshuffled at every epoch (default: False). sampler (Sampler, optional): defines the strategy to draw samples from the dataset. If specified, ``shuffle`` must be False. batch_sampler (Sampler, optional): like sampler, but returns a batch of indices at a time. Mutually exclusive with batch_size, shuffle, sampler, and drop_last. num_workers (int, optional): how many subprocesses to use for data loading. 0 means that the data will be loaded in the main process. (default: 0) collate_fn (callable, optional): merges a list of samples to form a mini-batch. pin_memory (bool, optional): If ``True``, the data loader will copy tensors into CUDA pinned memory before returning them. drop_last (bool, optional): set to ``True`` to drop the last incomplete batch, if the dataset size is not divisible by the batch size. If ``False`` and the size of dataset is not divisible by the batch size, then the last batch will be smaller. (default: False) timeout (numeric, optional): if positive, the timeout value for collecting a batch from workers. Should always be non-negative. (default: 0) worker_init_fn (callable, optional): If not None, this will be called on each worker subprocess with the worker id (an int in ``[0, num_workers - 1]``) as input, after seeding and before data loading. (default: None) .. note:: By default, each worker will have its PyTorch seed set to ``base_seed + worker_id``, where ``base_seed`` is a long generated by main process using its RNG. However, seeds for other libraies may be duplicated upon initializing workers (w.g., NumPy), causing each worker to return identical random numbers. (See :ref:`dataloader-workers-random-seed` section in FAQ.) You may use ``torch.initial_seed()`` to access the PyTorch seed for each worker in :attr:`worker_init_fn`, and use it to set other seeds before data loading. .. warning:: If ``spawn`` start method is used, :attr:`worker_init_fn` cannot be an unpicklable object, e.g., a lambda function. """</span> __initialized <span class="token operator">=</span> <span class="token boolean">False</span> <span class="token keyword">def</span> <span class="token function">__init__</span><span class="token punctuation">(</span>self<span class="token punctuation">,</span> dataset<span class="token punctuation">,</span> batch_size<span class="token operator">=</span><span class="token number">1</span><span class="token punctuation">,</span> shuffle<span class="token operator">=</span><span class="token boolean">False</span><span class="token punctuation">,</span> sampler<span class="token operator">=</span>None<span class="token punctuation">,</span> batch_sampler<span class="token operator">=</span>None<span class="token punctuation">,</span> num_workers<span class="token operator">=</span><span class="token number">0</span><span class="token punctuation">,</span> collate_fn<span class="token operator">=</span>default_collate<span class="token punctuation">,</span> pin_memory<span class="token operator">=</span><span class="token boolean">False</span><span class="token punctuation">,</span> drop_last<span class="token operator">=</span><span class="token boolean">False</span><span class="token punctuation">,</span> timeout<span class="token operator">=</span><span class="token number">0</span><span class="token punctuation">,</span> worker_init_fn<span class="token operator">=</span>None<span class="token punctuation">)</span><span class="token punctuation">:</span> self<span class="token punctuation">.</span>dataset <span class="token operator">=</span> dataset self<span class="token punctuation">.</span>batch_size <span class="token operator">=</span> batch_size self<span class="token punctuation">.</span>num_workers <span class="token operator">=</span> num_workers self<span class="token punctuation">.</span>collate_fn <span class="token operator">=</span> collate_fn self<span class="token punctuation">.</span>pin_memory <span class="token operator">=</span> pin_memory self<span class="token punctuation">.</span>drop_last <span class="token operator">=</span> drop_last self<span class="token punctuation">.</span>timeout <span class="token operator">=</span> timeout self<span class="token punctuation">.</span>worker_init_fn <span class="token operator">=</span> worker_init_fn <span class="token keyword">if</span> timeout <span class="token operator"><</span> <span class="token number">0</span><span class="token punctuation">:</span> <span class="token keyword">raise</span> ValueError<span class="token punctuation">(</span><span class="token string">'timeout option should be non-negative'</span><span class="token punctuation">)</span> <span class="token keyword">if</span> batch_sampler <span class="token keyword">is</span> <span class="token operator">not</span> None<span class="token punctuation">:</span> <span class="token keyword">if</span> batch_size <span class="token operator">></span> <span class="token number">1</span> <span class="token operator">or</span> shuffle <span class="token operator">or</span> sampler <span class="token keyword">is</span> <span class="token operator">not</span> None <span class="token operator">or</span> drop_last<span class="token punctuation">:</span> <span class="token keyword">raise</span> ValueError<span class="token punctuation">(</span><span class="token string">'batch_sampler option is mutually exclusive '</span> <span class="token string">'with batch_size, shuffle, sampler, and '</span> <span class="token string">'drop_last'</span><span class="token punctuation">)</span> self<span class="token punctuation">.</span>batch_size <span class="token operator">=</span> None self<span class="token punctuation">.</span>drop_last <span class="token operator">=</span> None <span class="token keyword">if</span> sampler <span class="token keyword">is</span> <span class="token operator">not</span> None <span class="token operator">and</span> shuffle<span class="token punctuation">:</span> <span class="token keyword">raise</span> ValueError<span class="token punctuation">(</span><span class="token string">'sampler option is mutually exclusive with '</span> <span class="token string">'shuffle'</span><span class="token punctuation">)</span> <span class="token keyword">if</span> self<span class="token punctuation">.</span>num_workers <span class="token operator"><</span> <span class="token number">0</span><span class="token punctuation">:</span> <span class="token keyword">raise</span> ValueError<span class="token punctuation">(</span><span class="token string">'num_workers option cannot be negative; '</span> <span class="token string">'use num_workers=0 to disable multiprocessing.'</span><span class="token punctuation">)</span> <span class="token keyword">if</span> batch_sampler <span class="token keyword">is</span> None<span class="token punctuation">:</span> <span class="token keyword">if</span> sampler <span class="token keyword">is</span> None<span class="token punctuation">:</span> <span class="token keyword">if</span> shuffle<span class="token punctuation">:</span> sampler <span class="token operator">=</span> RandomSampler<span class="token punctuation">(</span>dataset<span class="token punctuation">)</span> <span class="token operator">//</span>将list打乱 <span class="token keyword">else</span><span class="token punctuation">:</span> sampler <span class="token operator">=</span> SequentialSampler<span class="token punctuation">(</span>dataset<span class="token punctuation">)</span> batch_sampler <span class="token operator">=</span> BatchSampler<span class="token punctuation">(</span>sampler<span class="token punctuation">,</span> batch_size<span class="token punctuation">,</span> drop_last<span class="token punctuation">)</span> self<span class="token punctuation">.</span>sampler <span class="token operator">=</span> sampler self<span class="token punctuation">.</span>batch_sampler <span class="token operator">=</span> batch_sampler self<span class="token punctuation">.</span>__initialized <span class="token operator">=</span> <span class="token boolean">True</span> <span class="token keyword">def</span> <span class="token function">__setattr__</span><span class="token punctuation">(</span>self<span class="token punctuation">,</span> attr<span class="token punctuation">,</span> val<span class="token punctuation">)</span><span class="token punctuation">:</span> <span class="token keyword">if</span> self<span class="token punctuation">.</span>__initialized <span class="token operator">and</span> attr <span class="token keyword">in</span> <span class="token punctuation">(</span><span class="token string">'batch_size'</span><span class="token punctuation">,</span> <span class="token string">'sampler'</span><span class="token punctuation">,</span> <span class="token string">'drop_last'</span><span class="token punctuation">)</span><span class="token punctuation">:</span> <span class="token keyword">raise</span> ValueError<span class="token punctuation">(</span><span class="token string">'{} attribute should not be set after {} is '</span> <span class="token string">'initialized'</span><span class="token punctuation">.</span>format<span class="token punctuation">(</span>attr<span class="token punctuation">,</span> self<span class="token punctuation">.</span>__class__<span class="token punctuation">.</span>__name__<span class="token punctuation">)</span><span class="token punctuation">)</span> super<span class="token punctuation">(</span>DataLoader<span class="token punctuation">,</span> self<span class="token punctuation">)</span><span class="token punctuation">.</span>__setattr__<span class="token punctuation">(</span>attr<span class="token punctuation">,</span> val<span class="token punctuation">)</span> <span class="token keyword">def</span> <span class="token function">__iter__</span><span class="token punctuation">(</span>self<span class="token punctuation">)</span><span class="token punctuation">:</span> <span class="token keyword">return</span> _DataLoaderIter<span class="token punctuation">(</span>self<span class="token punctuation">)</span> <span class="token keyword">def</span> <span class="token function">__len__</span><span class="token punctuation">(</span>self<span class="token punctuation">)</span><span class="token punctuation">:</span> <span class="token keyword">return</span> len<span class="token punctuation">(</span>self<span class="token punctuation">.</span>batch_sampler<span class="token punctuation">)</span><span aria-hidden="true" class="line-numbers-rows"><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span></span></code></pre><pre class="line-numbers language-python"><code class="language-python">Arguments<span class="token punctuation">:</span> dataset<span class="token punctuation">(</span>Dataset<span class="token punctuation">)</span><span class="token punctuation">:</span> 传入的数据集 batch_size<span class="token punctuation">(</span>int<span class="token punctuation">,</span> optional<span class="token punctuation">)</span><span class="token punctuation">:</span> 每个batch有多少个样本 shuffle<span class="token punctuation">(</span>bool<span class="token punctuation">,</span> optional<span class="token punctuation">)</span><span class="token punctuation">:</span> 在每个epoch开始的时候,对数据进行重新排序 sampler<span class="token punctuation">(</span>Sampler<span class="token punctuation">,</span> optional<span class="token punctuation">)</span><span class="token punctuation">:</span> 自定义从数据集中取样本的策略,如果指定这个参数,那么shuffle必须为<span class="token boolean">False</span> batch_sampler<span class="token punctuation">(</span>Sampler<span class="token punctuation">,</span> optional<span class="token punctuation">)</span><span class="token punctuation">:</span> 与sampler类似,但是一次只返回一个batch的indices(索引),需要注意的是,一旦指定了这个参数,那么batch_size<span class="token punctuation">,</span>shuffle<span class="token punctuation">,</span>sampler<span class="token punctuation">,</span>drop_last就不能再制定了(互斥——Mutually exclusive) num_workers <span class="token punctuation">(</span>int<span class="token punctuation">,</span> optional<span class="token punctuation">)</span><span class="token punctuation">:</span> 这个参数决定了有几个进程来处理data loading。<span class="token number">0</span>意味着所有的数据都会被load进主进程。(默认为<span class="token number">0</span>) collate_fn <span class="token punctuation">(</span>callable<span class="token punctuation">,</span> optional<span class="token punctuation">)</span><span class="token punctuation">:</span> 将一个list的sample组成一个mini<span class="token operator">-</span>batch的函数 pin_memory <span class="token punctuation">(</span>bool<span class="token punctuation">,</span> optional<span class="token punctuation">)</span>: 如果设置为<span class="token boolean">True</span>,那么data loader将会在返回它们之前,将tensors拷贝到CUDA中的固定内存(CUDA pinned memory)中<span class="token punctuation">.</span> drop_last <span class="token punctuation">(</span>bool<span class="token punctuation">,</span> optional<span class="token punctuation">)</span><span class="token punctuation">:</span> 如果设置为<span class="token boolean">True</span>:这个是对最后的未完成的batch来说的,比如你的batch_size设置为<span class="token number">64</span>,而一个epoch只有<span class="token number">100</span>个样本,那么训练的时候后面的<span class="token number">36</span>个就被扔掉了…如果为<span class="token boolean">False</span>(默认),那么会继续正常执行,只是最后的batch_size会小一点。 timeout<span class="token punctuation">(</span>numeric<span class="token punctuation">,</span> optional<span class="token punctuation">)</span><span class="token punctuation">:</span> 如果是正数,表明等待从worker进程中收集一个batch等待的时间,若超出设定的时间还没有收集到,那就不收集这个内容了。这个numeric应总是大于等于<span class="token number">0</span>。默认为<span class="token number">0</span> worker_init_fn <span class="token punctuation">(</span>callable<span class="token punctuation">,</span> optional<span class="token punctuation">)</span><span class="token punctuation">:</span> 每个worker初始化函数<span class="token punctuation">,</span>默认为None。<span aria-hidden="true" class="line-numbers-rows"><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span></span></code></pre>]]></content>
<categories>
<category> Pytorch </category>
</categories>
<tags>
<tag> Learning Notes </tag>
</tags>
</entry>
<entry>
<title>Tag and Category的区别</title>
<link href="/posts/9d45.html"/>
<url>/posts/9d45.html</url>
<content type="html"><![CDATA[<h1 id="标签和分类的区别?"><a href="#标签和分类的区别?" class="headerlink" title="标签和分类的区别?"></a>标签和分类的区别?</h1><p>标签最显著的作用:</p><ul><li>一是传统意义上分类的作用,类似分类名称;</li><li>二是对文章内容进行一定程度的描述,类似于<strong>关键词</strong>。</li></ul><p>标签和分类还有一些细微的区别: </p><ul><li>同一篇文章标签可以用多个,但通常只能属于一个分类;</li><li>标签一般是在写作完成后,根据文章大意进行添加的;</li><li>标签可以把文章中重点词语提炼出来,有关键词的意义,但是分类没有;</li><li>标签通常反映了您的主要关注点,比如财经、互联网、体育等等;</li><li>标签给了我们一条联系他人的纽带,通过标签可以找到您的“同好”。</li></ul><p>Reference:</p><p>[1]<a href="https://iask.sina.com.cn/b/13948619.html">https://iask.sina.com.cn/b/13948619.html</a></p><p>Cover image by: <a href="https://huaban.com/pins/2618701579/">https://huaban.com/pins/2618701579/</a> </p>]]></content>
<categories>
<category> blog </category>
</categories>
<tags>
<tag> hexo </tag>
</tags>
</entry>
<entry>
<title>MICS新冠疫情在线讲座</title>
<link href="/posts/9dbf.html"/>
<url>/posts/9dbf.html</url>
<content type="html"><![CDATA[<h1 id="MICS新冠肺炎影像诊断及AI系统进展"><a href="#MICS新冠肺炎影像诊断及AI系统进展" class="headerlink" title="MICS新冠肺炎影像诊断及AI系统进展"></a>MICS新冠肺炎影像诊断及AI系统进展</h1><p><img src="https://cdn.jsdelivr.net/gh/sqzhang-jeremy/CDN/postimages/202002/20200221082755.png"></p><blockquote><p>武汉金银潭医院樊艳青主任曾说过一句:“我告诉自己不要流泪,眼睛要用来看片子”。放射科医生在疫情期间的工作压力可想而知。</p></blockquote><p>目前的疫情虽然说稍有缓和,但是一线的医生们仍面临着巨大的压力,此次MICS(Medical Imaging Computing Seminar )发起系列讲座,邀请一线医生、工业界、学术界的专家团队分别从各自视角围绕影像进行介绍和讨论。</p><p>此次lecture 由1一线视角:医生一线/病理/临床的视角介绍新冠肺炎、2AI医疗公司:联影智能、柏视医疗、沈睿医疗、推想科技、依图医疗结合疫情和本公司产品介绍算法如何在一线落地、3Panel环节:由医学界、学术界、工业界的专家们围绕讲座的内容和网友们提出的问题进行讨论,三个环节组成。</p><h2 id="1一线视角"><a href="#1一线视角" class="headerlink" title="1一线视角"></a>1一线视角</h2><p>目前一线诊断的方式:<strong>CT检查、核酸测试、临床诊断</strong>。</p><p>武汉协和医院放射科副主任史河水教授表示,**现在的工作强度是往年春节同期的3~4倍,而且每天肺部CT扫描占全天所有CT检查的90%**。同时,放射科医师可以通过制作教学课件、肺CT检查的科普读物,提高广大放射科医师对新冠肺炎CT征象的认识、老百姓对CT检查的了解,同样也是在为抗击疫情做贡献。武汉是肺炎疫情的核心区域,国际国内学术界也十分关注,希望武汉的医生能借助国内外的学术平台,及时交流经验,科学地抗击疫病。</p><blockquote><p>2月12日,新增新冠肺炎病例一下子上涨到了14840例。这是因为,湖北省将临床诊断病例数纳入了确诊病例数。根据第五版诊疗方案的表述,疑似病例具有肺炎影像学特征者,为临床诊断病例。新增案例的增长,从一定程度上说明了影像学在这次肺炎疫情中所发挥的价值。总的来说,新冠肺炎的一大特点是“快”。2月17日,国家卫健委医政医管局监察专员郭燕红表示,目前新冠肺炎患者从发病到确诊平均用时4.95天。</p></blockquote><h3 id="1-1病毒介绍"><a href="#1-1病毒介绍" class="headerlink" title="1.1病毒介绍"></a>1.1病毒介绍</h3><p><img src="https://cdn.jsdelivr.net/gh/sqzhang-jeremy/CDN/postimages/202002/20200221082800.png"></p><p> (就是这个病毒,原来长成这个鬼样子😡😡😡)</p><h3 id="1-2发病机制与病理"><a href="#1-2发病机制与病理" class="headerlink" title="1.2发病机制与病理"></a>1.2发病机制与病理</h3><p><img src="https://cdn.jsdelivr.net/gh/sqzhang-jeremy/CDN/postimages/202002/20200221090225.png"></p><h3 id="1-3临床分期及案例"><a href="#1-3临床分期及案例" class="headerlink" title="1.3临床分期及案例"></a>1.3临床分期及案例</h3><p><img src="https://cdn.jsdelivr.net/gh/sqzhang-jeremy/CDN/postimages/202002/20200221091226.png"></p><p><img src="https://cdn.jsdelivr.net/gh/sqzhang-jeremy/CDN/postimages/202002/20200221091242.png"></p><p><img src="https://cdn.jsdelivr.net/gh/sqzhang-jeremy/CDN/postimages/202002/20200221091244.png"></p><p><img src="https://cdn.jsdelivr.net/gh/sqzhang-jeremy/CDN/postimages/202002/20200221091245.png"></p><p><img src="https://cdn.jsdelivr.net/gh/sqzhang-jeremy/CDN/postimages/202002/20200221091243.png"></p><p> (SARS对比)</p><p><img src="https://cdn.jsdelivr.net/gh/sqzhang-jeremy/CDN/postimages/202002/20200221091247.png"></p><h3 id="1-4AI在2019-nCOV中的应用"><a href="#1-4AI在2019-nCOV中的应用" class="headerlink" title="1.4AI在2019-nCOV中的应用"></a>1.4AI在2019-nCOV中的应用</h3><p><img src="https://cdn.jsdelivr.net/gh/sqzhang-jeremy/CDN/postimages/202002/20200221091246.png"></p><p><img src="https://cdn.jsdelivr.net/gh/sqzhang-jeremy/CDN/postimages/202002/20200221091249.png"></p><blockquote><p>史河水主任表示,新冠肺炎患者肺内病变多,变化快,而且有些病变不易与其他感染或非感染性病变鉴别。医生凭借肉眼,要做出准确的定性诊断与定量评价,费时而且有难度。所以,很多医疗AI公司提出,在此前AI产品在肺部疾病解决方案的基础上,协助医生对新冠肺炎的诊断和疗效评价进行预估。总体来说,我个人认为AI在新冠肺炎的定量、定性评价方面,能提高医生的诊断效率。疫情爆发期过后,如何将散发新冠肺炎与其他病毒性肺炎等肺部病变进行鉴别,对放射科医生也是一种挑战,AI在这些方面也可以做一些工作。我们也需要这样一个工具,帮助我们更高效高质量地完成工作</p></blockquote><p><strong>目前,武汉协和医院的放射科室等各大医院在试用医疗AI公司的产品,主要的工作包括了病毒性肺炎在内的多种肺部疾病提供定性诊断、量化分析、多时点随访,标准结构化报告等辅助的智能诊断。</strong></p><p>目前,主流的AI产品大多是在影像方面通过机器学习、深度学习等技术针对病情结合标签进行训练预测病变位置、通过大数据寻找规律、可视化相关参数,来辅助医生进而提高诊断效率。由于数据方便的缘故,不仅仅由于数据少,而且结合潜伏期,CT扫描和核算测试多次鉴别的缘故,导致目前的系统完全达不到诊断效果,只能辅助。</p><h3 id="1-5总结"><a href="#1-5总结" class="headerlink" title="1.5总结"></a>1.5总结</h3><p><img src="https://cdn.jsdelivr.net/gh/sqzhang-jeremy/CDN/postimages/202002/20200221091248.png"></p>]]></content>
<categories>
<category> Lecture </category>
</categories>
<tags>
<tag> Talk </tag>
<tag> Medical Imaging </tag>
</tags>
</entry>
</search>