Skip to content

Commit

Permalink
modify logic regression
Browse files Browse the repository at this point in the history
  • Loading branch information
ad min committed Apr 14, 2016
1 parent 5a1e47f commit 0f06ce4
Showing 1 changed file with 1 addition and 1 deletion.
Original file line number Diff line number Diff line change
Expand Up @@ -46,7 +46,7 @@

  对于`k`类的多分类问题,模型的权重`w = (w_1, w_2, ..., w_{K-1})`是一个矩阵,如果添加截距,矩阵的维度为`(K-1) * (N+1)`,否则为`(K-1) * N`。单个样本的目标函数的损失函数可以写成如下公式**(7)**的形式。

<div align="center"><img src="imgs/2.2.png" width = "710" height = "110" alt="2.2" align="center" /></div><br>
<div align="center"><img src="imgs/2.2.png" width = "720" height = "165" alt="2.2" align="center" /></div><br>

&emsp;&emsp;对损失函数求一阶导数,我们可以得到下面的公式**(8)**:

Expand Down

0 comments on commit 0f06ce4

Please sign in to comment.