Machine Learning

开始: 2013-8-21

网易公开课: http://v.163.com/special/opencourse/machinelearning.html

对应资源网址:http://cs229.stanford.edu/materials.html

最新版本视频:https://class.coursera.org/ml-003/lecture/index (需要注册)

下载PPT的shell脚本:

 

#! /bin/sh

BASEURL="http://s3.amazonaws.com/mlclass-resources/docs/slides/Lecture"
SUFFIX=".pptx"

for i in `seq 1 1 18`
do
	fulllink=$BASEURL$i$SUFFIX
	echo "Downloading : $fulllink"
	wget $fulllink
done

exit 0

或者直接使用下面的下载地址:

http://s3.amazonaws.com/mlclass-resources/docs/slides/Lecture1.pptx
http://s3.amazonaws.com/mlclass-resources/docs/slides/Lecture2.pptx
http://s3.amazonaws.com/mlclass-resources/docs/slides/Lecture3.pptx
http://s3.amazonaws.com/mlclass-resources/docs/slides/Lecture4.pptx
http://s3.amazonaws.com/mlclass-resources/docs/slides/Lecture5.pptx
http://s3.amazonaws.com/mlclass-resources/docs/slides/Lecture6.pptx
http://s3.amazonaws.com/mlclass-resources/docs/slides/Lecture7.pptx
http://s3.amazonaws.com/mlclass-resources/docs/slides/Lecture8.pptx
http://s3.amazonaws.com/mlclass-resources/docs/slides/Lecture9.pptx
http://s3.amazonaws.com/mlclass-resources/docs/slides/Lecture10.pptx
http://s3.amazonaws.com/mlclass-resources/docs/slides/Lecture11.pptx
http://s3.amazonaws.com/mlclass-resources/docs/slides/Lecture12.pptx
http://s3.amazonaws.com/mlclass-resources/docs/slides/Lecture13.pptx
http://s3.amazonaws.com/mlclass-resources/docs/slides/Lecture14.pptx
http://s3.amazonaws.com/mlclass-resources/docs/slides/Lecture15.pptx
http://s3.amazonaws.com/mlclass-resources/docs/slides/Lecture16.pptx
http://s3.amazonaws.com/mlclass-resources/docs/slides/Lecture17.pptx
http://s3.amazonaws.com/mlclass-resources/docs/slides/Lecture18.pptx


学习心得:

第三课:关键是要理解 最大似然估计的思想。

第四课:GLM的3个假设不好理解。只是知道,利用GLM,只需要自己给输出变量y选择一个分布。然后按照那3个假设就可以求得假设函数h(x)。求解参数,同样利用最大似然估计,结合梯度上升算法,或牛顿方法,都可以求得参数。

其他学习资料: http://www.cnblogs.com/tornadomeet/archive/2013/03/14/2959138.html 记录Deep learning的知识,很详细

第五课:理解贝叶斯定理即可。



==================其他人整理的机器学习资料=================

http://blog.csdn.net/lefter1986/article/details/9842519

你可能感兴趣的:(Machine Learning)