🎉 修改格式

Relph1119 · Oct 14, 2023 · 9d26bf4 · 9d26bf4
1 parent 7105f4b
commit 9d26bf4
Show file tree

Hide file tree

Showing 2 changed files with 18 additions and 18 deletions.
diff --git a/docs/diffusion_models_learning51/ch01.md b/docs/diffusion_models_learning51/ch01.md
@@ -22,24 +22,24 @@ $$
 - 反向过程（数据去噪化）：从采样自高斯噪声$x \sim \mathcal{N}(0, I)$的随机噪声中恢复出原始数据$x_0$，通过由一系列用神经网络参数化的高斯分布组成的马尔可夫链进行数据去噪。
 
     从时间步$t$到时间步$t - 1$的单步反向“去噪”过程：
-    $$
+$$
 q(x_{t - 1} | x_t, x_0) = \mathcal{N} (x_{t - 1}; \tilde{\mu}(x_t, x_0), \tilde{\beta_t} I)    
-    $$
-    其中，均值为
-    $$
+$$
+&emsp;&emsp;其中，均值为
+$$
 \tilde{\beta_t} = \frac{ 1 - \bar{\alpha}_{t - 1}}{ 1 - \bar{\alpha}_t} \cdot \beta_t
-    $$
-    方差为
-    $$
+$$
+&emsp;&emsp;方差为
+$$
 \tilde{\mu}(x_t, x_0) = \frac{\sqrt{\alpha_t} (1 - \bar{\alpha}_{t - 1})}{1 - \bar{\alpha}_t} x_t + \frac{ \sqrt{ \bar{\alpha}_{t - 1} } \beta_t }{1 - \bar{\alpha}_t} x_0   
-    $$
+$$
 
 - 优化目标：要求反向过程中预测的噪声分布与前向过程中施加的噪声分布之间的“距离”最小。
 
     优化目标公式：
-    $$
+$$
 L_{t - 1}^{\text{simple}} = \mathbb{E}_{x_0, \epsilon \sim \mathcal{N}(0, I)} \left[ \| \epsilon - \epsilon_{\theta} \left( \sqrt{\bar{\alpha}_t} x_0 + \sqrt{1 - \bar{\alpha}_t} \epsilon, t \right) \|^2  \right]    
-    $$
+$$
 
 ## 1.2 扩散模型的发展
 

diff --git a/notebook/diffusion_models_learning51/my_homework/ch01.ipynb b/notebook/diffusion_models_learning51/my_homework/ch01.ipynb
@@ -71,17 +71,17 @@
     "- 反向过程（数据去噪化）：从采样自高斯噪声$x \\sim \\mathcal{N}(0, I)$的随机噪声中恢复出原始数据$x_0$，通过由一系列用神经网络参数化的高斯分布组成的马尔可夫链进行数据去噪。\n",
     "\n",
     "    从时间步$t$到时间步$t - 1$的单步反向“去噪”过程：\n",
-    "    $$\n",
+    "$$\n",
     "q(x_{t - 1} | x_t, x_0) = \\mathcal{N} (x_{t - 1}; \\tilde{\\mu}(x_t, x_0), \\tilde{\\beta_t} I)    \n",
-    "    $$\n",
+    "$$\n",
     "    其中，均值为\n",
-    "    $$\n",
+    "$$\n",
     "\\tilde{\\beta_t} = \\frac{ 1 - \\bar{\\alpha}_{t - 1}}{ 1 - \\bar{\\alpha}_t} \\cdot \\beta_t\n",
-    "    $$\n",
+    "$$\n",
     "    方差为\n",
-    "    $$\n",
+    "$$\n",
     "\\tilde{\\mu}(x_t, x_0) = \\frac{\\sqrt{\\alpha_t} (1 - \\bar{\\alpha}_{t - 1})}{1 - \\bar{\\alpha}_t} x_t + \\frac{ \\sqrt{ \\bar{\\alpha}_{t - 1} } \\beta_t }{1 - \\bar{\\alpha}_t} x_0   \n",
-    "    $$"
+    "$$"
    ]
   },
   {
@@ -92,9 +92,9 @@
     "- 优化目标：要求反向过程中预测的噪声分布与前向过程中施加的噪声分布之间的“距离”最小。\n",
     "\n",
     "    优化目标公式：\n",
-    "    $$\n",
+    "$$\n",
     "L_{t - 1}^{\\text{simple}} = \\mathbb{E}_{x_0, \\epsilon \\sim \\mathcal{N}(0, I)} \\left[ \\| \\epsilon - \\epsilon_{\\theta} \\left( \\sqrt{\\bar{\\alpha}_t} x_0 + \\sqrt{1 - \\bar{\\alpha}_t} \\epsilon, t \\right) \\|^2  \\right]    \n",
-    "    $$"
+    "$$"
    ]
   },
   {