Skip to content

Commit

Permalink
update some chapter 2 pictures
Browse files Browse the repository at this point in the history
  • Loading branch information
hwade committed Nov 16, 2018
1 parent 701d764 commit ce58ba0
Show file tree
Hide file tree
Showing 16 changed files with 9 additions and 7 deletions.
6 changes: 4 additions & 2 deletions .gitignore
Original file line number Diff line number Diff line change
@@ -1,2 +1,4 @@
.ipynb_checkpoints
.DS_Store
.ipynb_checkpoints
.DS_Store
.gitignore
*.ipynb
Binary file modified ch02_机器学习基础/img/ch2/2.1/1.jpg
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Binary file modified ch02_机器学习基础/img/ch2/2.1/11.jpg
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Binary file modified ch02_机器学习基础/img/ch2/2.1/2.jpg
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Binary file added ch02_机器学习基础/img/ch2/2.1/3.jpg
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Binary file removed ch02_机器学习基础/img/ch2/2.1/3.png
Binary file not shown.
Binary file modified ch02_机器学习基础/img/ch2/2.1/5.jpg
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Binary file added ch02_机器学习基础/img/ch2/2.1/7.jpg
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Binary file removed ch02_机器学习基础/img/ch2/2.1/7.png
Binary file not shown.
Binary file added ch02_机器学习基础/img/ch2/2.16/1.jpg
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Binary file removed ch02_机器学习基础/img/ch2/2.16/1.png
Binary file not shown.
Binary file added ch02_机器学习基础/img/ch2/2.16/2.jpg
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Binary file removed ch02_机器学习基础/img/ch2/2.16/2.png
Binary file not shown.
Binary file added ch02_机器学习基础/img/ch2/2.18/1.jpg
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Binary file removed ch02_机器学习基础/img/ch2/2.18/1.png
Binary file not shown.
10 changes: 5 additions & 5 deletions ch02_机器学习基础/第二章_机器学习基础.md
Original file line number Diff line number Diff line change
Expand Up @@ -8,15 +8,15 @@

|回归算法|基于实例的算法|正则化方法|
|:-:|:-:|:-:|
|![](./img/ch2/2.1/1.jpg)|![](./img/ch2/2.1/2.jpg)|![](./img/ch2/2.1/3.png)|
|![](./img/ch2/2.1/1.jpg)|![](./img/ch2/2.1/2.jpg)|![](./img/ch2/2.1/3.jpg)|

|决策树学习|贝叶斯方法|基于核的算法|
|:-:|:-:|:-:|
|![](./img/ch2/2.1/4.png)|![](./img/ch2/2.1/5.jpg)|![](./img/ch2/2.1/6.jpg)|

|聚类算法|关联规则学习|人工神经网络|
|:-:|:-:|:-:|
|![](./img/ch2/2.1/7.png)|![](./img/ch2/2.1/8.jpg)|![](./img/ch2/2.1/9.png)|
|![](./img/ch2/2.1/7.jpg)|![](./img/ch2/2.1/8.jpg)|![](./img/ch2/2.1/9.png)|

|深度学习|降低维度算法|集成算法|
|:-:|:-:|:-:|
Expand Down Expand Up @@ -239,7 +239,7 @@ $$
假设函数中有$A$和$B$两个参数,当参数发生变化时,假设函数状态也会随着变化。
如下图所示

![](./img/ch2/2.16/1.png)
![](./img/ch2/2.16/1.jpg)

想要你和图中的离散点,我们需要尽可能找到最优的$A$和$B$来使这条直线更能代表所有数据。如何找到最优解呢,这就需要使用代价函数来求解,以平方误差代价函数为例,假设函数为$h(x)=\theta_0x$。
平方误差代价函数的主要思想
Expand All @@ -253,7 +253,7 @@ $$
**最优解即为代价函数的最小值**$\min J(\theta_0, \theta_1)$。如果是1个参数,代价函数一般通过二维曲线便可直观看出。如果是2个参数,代价函数通过三维图像可看出效果,参数越多,越复杂。
当参数为2个时,代价函数是三维图像。

![](./img/ch2/2.16/2.png)
![](./img/ch2/2.16/2.jpg)

### 2.10.3 为什么代价函数要非负?
目标函数存在一个下界,在优化过程当中,如果优化算法能够使目标函数不断减小,根据单调有界准则,这个优化算法就能证明是收敛有效的。
Expand All @@ -278,7 +278,7 @@ $$\frac{\delta J}{\delta w}=(a-y)\delta'(z)x$$,$$\frac{\delta J}{\delta b}=(a-

**:神经网络常用的激活函数为sigmoid函数,该函数的曲线如下所示:

![](./img/ch2/2.18/1.png)
![](./img/ch2/2.18/1.jpg)

假设目标是收敛到1.0。0.82离目标比较远,梯度比较大,权值调整比较大。0.98离目标比较近,梯度比较小,权值调整比较小。调整方案合理。
假如目标是收敛到0。0.82目标比较近,梯度比较大,权值调整比较大。0.98离目标比较远,梯度比较小,权值调整比较小。调整方案不合理。
Expand Down

0 comments on commit ce58ba0

Please sign in to comment.