Add new post and adjust image path

jacksonwuu · Feb 8, 2024 · 4738996 · 4738996
1 parent a50c678
commit 4738996
Show file tree

Hide file tree

Showing 28 changed files with 40 additions and 23 deletions.
diff --git a/_config.yml b/_config.yml
@@ -43,7 +43,7 @@ external_link:
   exclude: ''
 filename_case: 0
 render_drafts: false
-post_asset_folder: false
+post_asset_folder: true
 relative_link: false
 future: true
 syntax_highlighter: highlight.js

diff --git a/package-lock.json b/package-lock.json
diff --git a/package.json b/package.json
@@ -18,6 +18,7 @@
     "hexo-generator-feed": "^3.0.0",
     "hexo-generator-index": "^3.0.0",
     "hexo-generator-tag": "^2.0.0",
+    "hexo-image-link": "^0.0.5",
     "hexo-renderer-ejs": "^2.0.0",
     "hexo-renderer-marked": "^6.0.0",
     "hexo-renderer-stylus": "^3.0.0",

diff --git a/source/_posts/2023-04-deep-learning-notes.md b/source/_posts/2023-04-deep-learning-notes.md
@@ -28,9 +28,9 @@ PPMI 矩阵是对共现矩阵的一种改进。但 PPMI 矩阵还是存在一个
 
 降维的方法有很多，这本书上用到的是奇异值分解(Singular Value Decomposition，SVD)。 这一块是线性代数的知识，可以去看看 3b1b 的线性代数教程，通过可视化的方式更好地理解线性代数。
 
-![](/images/2023-04-deep-learning-notes/%E4%B8%8A%E4%BD%8D%E4%B8%8B%E4%BD%8D%E5%85%B3%E7%B3%BB%E5%9B%BE.png)
+![](2023-04-deep-learning-notes/%E4%B8%8A%E4%BD%8D%E4%B8%8B%E4%BD%8D%E5%85%B3%E7%B3%BB%E5%9B%BE.png)
 
-![](/images/2023-04-deep-learning-notes/%E9%99%8D%E7%BB%B4%E7%A4%BA%E6%84%8F%E5%9B%BE.png)
+![](2023-04-deep-learning-notes/%E9%99%8D%E7%BB%B4%E7%A4%BA%E6%84%8F%E5%9B%BE.png)
 
 ## 第三四章
 
@@ -40,9 +40,9 @@ PPMI 矩阵是对共现矩阵的一种改进。但 PPMI 矩阵还是存在一个
 
 这一章主要讲了 word2vec 与其 CBOW 模型。
 
-![](/images/2023-04-deep-learning-notes/CBOW%E6%A8%A1%E5%9E%8B.png)
+![](2023-04-deep-learning-notes/CBOW%E6%A8%A1%E5%9E%8B.png)
 
-![](/images/2023-04-deep-learning-notes/%E5%BE%AE%E7%8E%87%E5%88%86%E5%B8%83.png)
+![](2023-04-deep-learning-notes/%E5%BE%AE%E7%8E%87%E5%88%86%E5%B8%83.png)
 
 ## 第四章
 
@@ -58,11 +58,11 @@ PPMI 矩阵是对共现矩阵的一种改进。但 PPMI 矩阵还是存在一个
 
 这一章剩下的部分基本上都在讲高数（微分、导数、偏导、梯度），下一次打卡再总结。
 
-![](/images/2023-04-deep-learning-notes/Input-Data.webp)
+![](2023-04-deep-learning-notes/Input-Data.webp)
 
-![](/images/2023-04-deep-learning-notes/NIPTHOG9.png)
+![](2023-04-deep-learning-notes/NIPTHOG9.png)
 
-![](/images/2023-04-deep-learning-notes/Upperloop.png)
+![](2023-04-deep-learning-notes/Upperloop.png)
 
 ## 第五章
 
@@ -85,11 +85,11 @@ PPMI 矩阵是对共现矩阵的一种改进。但 PPMI 矩阵还是存在一个
 
 理解了反向传播算法后，整个实现也是非常容易理解的。
 
-![](/images/2023-04-deep-learning-notes/Backprspaoation.png)
+![](2023-04-deep-learning-notes/Backprspaoation.png)
 
-![](/images/2023-04-deep-learning-notes/%E5%8F%8D%E5%90%91%E4%BC%A0%E6%92%AD.png)
+![](2023-04-deep-learning-notes/%E5%8F%8D%E5%90%91%E4%BC%A0%E6%92%AD.png)
 
-![](/images/2023-04-deep-learning-notes/%E7%A5%9E%E7%BB%8F%E7%BD%91%E7%BB%9C%E5%85%A8%E8%B2%8C%E5%9B%BE.png)
+![](2023-04-deep-learning-notes/%E7%A5%9E%E7%BB%8F%E7%BD%91%E7%BB%9C%E5%85%A8%E8%B2%8C%E5%9B%BE.png)
 
 ## 第六章
 
@@ -119,11 +119,11 @@ SGD 的缺点是如果函数的形状非均向(anisotropic)，比如呈延伸状
 
 这一章后面的内容等到下次打卡。
 
-![](/images/2023-04-deep-learning-notes/%E6%96%B9%E6%B3%95%E6%AF%94%E8%BE%83.png)
+![](2023-04-deep-learning-notes/%E6%96%B9%E6%B3%95%E6%AF%94%E8%BE%83.png)
 
-![](/images/2023-04-deep-learning-notes/MNIST.png)
+![](2023-04-deep-learning-notes/MNIST.png)
 
-![](/images/2023-04-deep-learning-notes/Algorithms.png)
+![](2023-04-deep-learning-notes/Algorithms.png)
 
 ## 继续打卡第六章
 
@@ -145,9 +145,9 @@ Dropout 是一种在学习的过程中随机删除神经元的方法。训练时
 
 这一章有很多内容还不够理解，值得后续再研究一番。
 
-![](/images/2023-04-deep-learning-notes/Without.webp)
+![](2023-04-deep-learning-notes/Without.webp)
 
-![](/images/2023-04-deep-learning-notes/%E8%B6%85%E5%8F%82%E6%95%B0%E4%BC%98%E5%8C%96.png)
+![](2023-04-deep-learning-notes/%E8%B6%85%E5%8F%82%E6%95%B0%E4%BC%98%E5%8C%96.png)
 
 ## 打卡第七章——卷积神经网络（CNN）！！！
 
@@ -167,14 +167,14 @@ CNN 中有时将卷积层的输入输出数据称为特征图(feature map)。
 
 具有代表性的 CNN 有 LeNet 和 AlexNet。LeNet 和 AlexNet 没有太大的不同。但围绕它们的环境和计算机技术有了很大的进步。具体地说，现在任何人都可以获得大量的数据。而且，擅长大规模并行计算的 GPU 得到普及，高速进行大量的运算已经成为可能。大数据和 GPU 已成为深度学习发展的巨大的原动力！
 
-![](/images/2023-04-deep-learning-notes/CNN.png)
+![](2023-04-deep-learning-notes/CNN.png)
 
-![](/images/2023-04-deep-learning-notes/RGB.png)
+![](2023-04-deep-learning-notes/RGB.png)
 
-![](/images/2023-04-deep-learning-notes/%E5%8D%B7%E7%A7%AF%E8%BF%90%E7%AE%97.png)
+![](2023-04-deep-learning-notes/%E5%8D%B7%E7%A7%AF%E8%BF%90%E7%AE%97.png)
 
-![](/images/2023-04-deep-learning-notes/url.jpg)
+![](2023-04-deep-learning-notes/url.jpg)
 
-![](/images/2023-04-deep-learning-notes/7-26.png)
+![](2023-04-deep-learning-notes/7-26.png)
 
-![](/images/2023-04-deep-learning-notes/Feature-map.webp)
+![](2023-04-deep-learning-notes/Feature-map.webp)
diff --git a/...ages/2023-04-deep-learning-notes/7-26.png → ...osts/2023-04-deep-learning-notes/7-26.png b/...ages/2023-04-deep-learning-notes/7-26.png → ...osts/2023-04-deep-learning-notes/7-26.png
diff --git a/...023-04-deep-learning-notes/Algorithms.png → ...023-04-deep-learning-notes/Algorithms.png b/...023-04-deep-learning-notes/Algorithms.png → ...023-04-deep-learning-notes/Algorithms.png
diff --git a/...4-deep-learning-notes/Backprspaoation.png → ...4-deep-learning-notes/Backprspaoation.png b/...4-deep-learning-notes/Backprspaoation.png → ...4-deep-learning-notes/Backprspaoation.png
diff --git a/...es/2023-04-deep-learning-notes/CBOW模型.png → ...ts/2023-04-deep-learning-notes/CBOW模型.png b/...es/2023-04-deep-learning-notes/CBOW模型.png → ...ts/2023-04-deep-learning-notes/CBOW模型.png
diff --git a/...mages/2023-04-deep-learning-notes/CNN.png → ...posts/2023-04-deep-learning-notes/CNN.png b/...mages/2023-04-deep-learning-notes/CNN.png → ...posts/2023-04-deep-learning-notes/CNN.png
diff --git a/...3-04-deep-learning-notes/Feature-map.webp → ...3-04-deep-learning-notes/Feature-map.webp b/...3-04-deep-learning-notes/Feature-map.webp → ...3-04-deep-learning-notes/Feature-map.webp
diff --git a/...23-04-deep-learning-notes/Input-Data.webp → ...23-04-deep-learning-notes/Input-Data.webp b/...23-04-deep-learning-notes/Input-Data.webp → ...23-04-deep-learning-notes/Input-Data.webp
diff --git a/...ges/2023-04-deep-learning-notes/MNIST.png → ...sts/2023-04-deep-learning-notes/MNIST.png b/...ges/2023-04-deep-learning-notes/MNIST.png → ...sts/2023-04-deep-learning-notes/MNIST.png
diff --git a/.../2023-04-deep-learning-notes/NIPTHOG9.png → .../2023-04-deep-learning-notes/NIPTHOG9.png b/.../2023-04-deep-learning-notes/NIPTHOG9.png → .../2023-04-deep-learning-notes/NIPTHOG9.png
diff --git a/...mages/2023-04-deep-learning-notes/RGB.png → ...posts/2023-04-deep-learning-notes/RGB.png b/...mages/2023-04-deep-learning-notes/RGB.png → ...posts/2023-04-deep-learning-notes/RGB.png
diff --git a/...2023-04-deep-learning-notes/Upperloop.png → ...2023-04-deep-learning-notes/Upperloop.png b/...2023-04-deep-learning-notes/Upperloop.png → ...2023-04-deep-learning-notes/Upperloop.png
diff --git a/.../2023-04-deep-learning-notes/Without.webp → .../2023-04-deep-learning-notes/Without.webp b/.../2023-04-deep-learning-notes/Without.webp → .../2023-04-deep-learning-notes/Without.webp
diff --git a/...mages/2023-04-deep-learning-notes/url.jpg → ...posts/2023-04-deep-learning-notes/url.jpg b/...mages/2023-04-deep-learning-notes/url.jpg → ...posts/2023-04-deep-learning-notes/url.jpg
diff --git a/...s/2023-04-deep-learning-notes/上位下位关系图.png → ...s/2023-04-deep-learning-notes/上位下位关系图.png b/...s/2023-04-deep-learning-notes/上位下位关系图.png → ...s/2023-04-deep-learning-notes/上位下位关系图.png
diff --git a/...ages/2023-04-deep-learning-notes/卷积运算.png → ...osts/2023-04-deep-learning-notes/卷积运算.png b/...ages/2023-04-deep-learning-notes/卷积运算.png → ...osts/2023-04-deep-learning-notes/卷积运算.png
diff --git a/...ages/2023-04-deep-learning-notes/反向传播.png → ...osts/2023-04-deep-learning-notes/反向传播.png b/...ages/2023-04-deep-learning-notes/反向传播.png → ...osts/2023-04-deep-learning-notes/反向传播.png
diff --git a/...ages/2023-04-deep-learning-notes/微率分布.png → ...osts/2023-04-deep-learning-notes/微率分布.png b/...ages/2023-04-deep-learning-notes/微率分布.png → ...osts/2023-04-deep-learning-notes/微率分布.png
diff --git a/...ages/2023-04-deep-learning-notes/方法比较.png → ...osts/2023-04-deep-learning-notes/方法比较.png b/...ages/2023-04-deep-learning-notes/方法比较.png → ...osts/2023-04-deep-learning-notes/方法比较.png
diff --git a/...s/2023-04-deep-learning-notes/神经网络全貌图.png → ...s/2023-04-deep-learning-notes/神经网络全貌图.png b/...s/2023-04-deep-learning-notes/神经网络全貌图.png → ...s/2023-04-deep-learning-notes/神经网络全貌图.png
diff --git a/...ges/2023-04-deep-learning-notes/超参数优化.png → ...sts/2023-04-deep-learning-notes/超参数优化.png b/...ges/2023-04-deep-learning-notes/超参数优化.png → ...sts/2023-04-deep-learning-notes/超参数优化.png
diff --git a/...ges/2023-04-deep-learning-notes/降维示意图.png → ...sts/2023-04-deep-learning-notes/降维示意图.png b/...ges/2023-04-deep-learning-notes/降维示意图.png → ...sts/2023-04-deep-learning-notes/降维示意图.png
diff --git a/source/_posts/2024-02-kbuild.md b/source/_posts/2024-02-kbuild.md
@@ -46,7 +46,7 @@ Linux kernel 正是有了 kbuild 这样强大的构建系统，它才能很灵
 
 执行这个命令之后，它会在命令行中出现一个配置界面，通过这个配置界面，用户可以勾选各种配置参数，然后点击保存，它就会生成`.config`这个文件，这个文件就是编译整个内核的配置文件。
 
-![](2024-02-08-11-04-32.png)
+![](2024-02-kbuild/2024-02-08-11-04-32.png)
 
 执行`make menuconfig` 会触发根目录下的`Makefile`里的`%config` 这个 target：
 

diff --git a/source/_posts/2024-02-08-11-04-32.png → ...ts/2024-02-kbuild/2024-02-08-11-04-32.png b/source/_posts/2024-02-08-11-04-32.png → ...ts/2024-02-kbuild/2024-02-08-11-04-32.png
diff --git a/yarn.lock b/yarn.lock
@@ -648,6 +648,11 @@
   dependencies:
     "sprintf-js" "^1.1.2"
 
+"hexo-image-link@^0.0.5":
+  "integrity" "sha512-IakmAA0lPATF6iUfHAzYSuZPA6DYLlMj4LTGnT4LeEtmIHjPWsTDgGssPbMJakA47M1vNv5WwiNS3GCtdH2FQw=="
+  "resolved" "https://registry.npmjs.org/hexo-image-link/-/hexo-image-link-0.0.5.tgz"
+  "version" "0.0.5"
+
 "hexo-log@^4.0.1":
   "integrity" "sha512-i2Sgxk8Cgx5viSjq5qW5N/rBFfwoCKQcH8qnnW1fawCapcdEAhIsq+Y3vbrs9bssyDlyU6Vqm4oQmosREaNI7Q=="
   "resolved" "https://registry.npmjs.org/hexo-log/-/hexo-log-4.1.0.tgz"