Added missing log function on week 7. (#826)

* [ar] Added missing log function * [es] Added missing log function * [es] Added missing log * [FR] Added missing log function. * [IT] Added missing log function. * [PT] Added missing log function. * [KO] Added missing log function. * [TR] Added missing log function. * [ZH] Added missing log function. * [EN] Added missing log function. Look at the lecture videos presentation.
Atcold · Aug 9, 2022 · c42d8e7 · c42d8e7
1 parent 5ad417c
commit c42d8e7
Show file tree

Hide file tree

Showing 9 changed files with 10 additions and 10 deletions.
diff --git a/docs/ar/week07/07-2.md b/docs/ar/week07/07-2.md
@@ -67,7 +67,7 @@ $$
 أقصى احتمال يحاول جعل البسط كبيرًا والمقام صغيرًا لزيادة الاحتمالية. هذا يعادل تقليص $-\log(P(Y \mid W))$ كما هو موضح أدناه
 
 $$
-L(Y, W) = E(Y,W) + \frac{1}{\beta}\int_{y}e^{-\beta E(y,W)}
+L(Y, W) = E(Y,W) + \frac{1}{\beta}\log\int_{y}e^{-\beta E(y,W)}
 $$
 
 حساب الانحدار لخسارة احتمالية اللوغاريتم السلبي لعينة واحدة من Y  يكون كالتالي:

diff --git a/docs/en/week07/07-2.md b/docs/en/week07/07-2.md
@@ -67,7 +67,7 @@ $$
 Maximum likelihood tries to make the numerator big and the denominator small to maximize the likelihood. This is equivalent to minimizing $-\log(P(Y \mid W))$ which is given below
 
 $$
-L(Y, W) = E(Y,W) + \frac{1}{\beta}\int_{y}e^{-\beta E(y,W)}
+L(Y, W) = E(Y,W) + \frac{1}{\beta}\log\int_{y}e^{-\beta E(y,W)}
 $$
 
 Gradient of the negative log likelihood loss for one sample Y is as follows:

diff --git a/docs/es/week07/07-2.md b/docs/es/week07/07-2.md
@@ -122,7 +122,7 @@ $$
 Maximum likelihood tries to make the numerator big and the denominator small to maximize the likelihood. This is equivalent to minimizing $-\log(P(Y \mid W))$ which is given below
 
 $$
-L(Y, W) = E(Y,W) + \frac{1}{\beta}\int_{y}e^{-\beta E(y,W)}
+L(Y, W) = E(Y,W) + \frac{1}{\beta} \log \int_{y}e^{-\beta E(y,W)}
 $$
 
 Gradient of the negative log likelihood loss for one sample Y is as follows:
@@ -154,7 +154,7 @@ La verosimilitud máxima intenta hacer que el numerador sea grande y el denomina
 
 
 $$
-L(Y, W) = E(Y,W) + \frac{1}{\beta}\int_{y}e^{-\beta E(y,W)}
+L(Y, W) = E(Y,W) + \frac{1}{\beta}\log\int_{y}e^{-\beta E(y,W)}
 $$
 
 El gradiente de la funcion de pérdida de probabilidad logarítmica negativa para una muestra Y es el siguiente:

diff --git a/docs/fr/week07/07-2.md b/docs/fr/week07/07-2.md
@@ -145,7 +145,7 @@ $$
 Le maximum de vraisemblance essaie de rendre le numérateur grand et le dénominateur petit pour maximiser la probabilité. Cela équivaut à minimiser $-\log(P(Y \mid W))$ qui est donné ci-dessous :
 
 $$
-L(Y, W) = E(Y,W) + \frac{1}{\beta}\int_{y}e^{-\beta E(y,W)}
+L(Y, W) = E(Y,W) + \frac{1}{\beta}\log\int_{y}e^{-\beta E(y,W)}
 $$
 
 Le gradient de la perte de la vraisemblance logarithmique négative pour un échantillon Y est le suivant :

diff --git a/docs/it/week07/07-2.md b/docs/it/week07/07-2.md
@@ -145,7 +145,7 @@ Maximum likelihood tries to make the numerator big and the denominator small to
 -->
 
 $$
-L(Y, W) = E(Y,W) + \frac{1}{\beta}\int_{y}e^{-\beta E(y,W)}
+L(Y, W) = E(Y,W) + \frac{1}{\beta}\log\int_{y}e^{-\beta E(y,W)}
 $$
 
 Gradient of the negative log likelihoood loss for one sample Y is as follows:

diff --git a/docs/ko/week07/07-2.md b/docs/ko/week07/07-2.md
@@ -115,7 +115,7 @@ $$
 최대 우도는 분자를 크게하고, 분모를 작게해서 우도 값을 최대화 하고자 한다. 이는 아래에서 주어진 것과 같이 $-\log(P(Y \mid W))$ 을 최소화 하는 것과 같다. 
 
 $$
-L(Y, W) = E(Y,W) + \frac{1}{\beta}\int_{y}e^{-\beta E(y,W)}
+L(Y, W) = E(Y,W) + \frac{1}{\beta}\log\int_{y}e^{-\beta E(y,W)}
 $$
 
 <!-- Gradient of the negative log likelihoood loss for one sample Y is as follows: -->

diff --git a/docs/pt/week07/07-2.md b/docs/pt/week07/07-2.md
@@ -128,7 +128,7 @@ $$
 A máxima verossimilhança tenta tornar o numerador grande e o denominador pequeno para maximizar a verossimilhança. Isso é equivalente a minimizar $-\log(P(Y \mid W))$ que é dado abaixo
 
 <!--$$
-L(Y, W) = E(Y,W) + \frac{1}{\beta}\int_{y}e^{-\beta E(y,W)}
+L(Y, W) = E(Y,W) + \frac{1}{\beta}\log\int_{y}e^{-\beta E(y,W)}
 $$
 -->
 

diff --git a/docs/tr/week07/07-2.md b/docs/tr/week07/07-2.md
@@ -122,7 +122,7 @@ En Büyük Olabilirlik, olabilirliği maksimuma çıkarmak için payı büyütü
 <!--Maximum likelihood tries to make the numerator big and the denominator small to maximize the likelihood. This is equivalent to minimizing $-\log(P(Y \mid W))$ which is given below-->
 
 $$
-L(Y, W) = E(Y,W) + \frac{1}{\beta}\int_{y}e^{-\beta E(y,W)}
+L(Y, W) = E(Y,W) + \frac{1}{\beta}\log\int_{y}e^{-\beta E(y,W)}
 $$
 
 Bir Y örneği için negatif log olabilirlik kaybının gradyanı aşağıdaki gibi:

diff --git a/docs/zh/week07/07-2.md b/docs/zh/week07/07-2.md
@@ -70,7 +70,7 @@ $$
 最大似然尝试令分子变大，而分母变细来最大化似然。这是跟最小化 $-\log(P(Y \mid W))$ 一样，也就是以下这样
 
 $$
-L(Y, W) = E(Y,W) + \frac{1}{\beta}\int_{y}e^{-\beta E(y,W)}
+L(Y, W) = E(Y,W) + \frac{1}{\beta}\log\int_{y}e^{-\beta E(y,W)}
 $$
 
 对于一个样本 Y ，负对数似然损失的的梯度(Gradient of the negative log likelihood loss)是如下:
-Original file line number
+Diff line change
@@ Expand Up @@
     -->
     $$
-    L(Y, W) = E(Y,W) + \frac{1}{\beta}\int_{y}e^{-\beta E(y,W)}
+    L(Y, W) = E(Y,W) + \frac{1}{\beta}\log\int_{y}e^{-\beta E(y,W)}
     $$
     Gradient of the negative log likelihoood loss for one sample Y is as follows:
@@ Expand Down @@