约翰霍普金斯大学 data science toolbox 笔记

data scientists toolbox

这门课是由生物学院的老师讲授的，选择的是R语言
statistics is the science of learning from data
需要学会用google和stackoverflow, ie 学会一些hacking skill
列出了这些参考书Specialization Textbooks

Elements of Data Analytic Style by Jeff Leek
R Programming for Data Science by Roger Peng
Exploratory Data Analysis with R by Roger Peng
Report Writing for Data Science in R by Roger Peng
Statistical Inference for Data Science by Brian Caffo
Regression Modeling for Data Science in R by Brian Caffo
Developing Data Products in R by Brian Caffo
In addition, to the above books, two additional books that are highly relevant to the Specialization are

The Art of Data Science by Roger Peng
How to Be A Modern Scientist by Jeff Leek
配套参考书
the elements of data science 电子版下了一本 98 页
另外有个更全的 相关书籍的整理
安装r语言，安装interface Rstudio
安装git，安装github
基本的命令行，基本的git command
基本的hack skill ，比如说如何google/stack overflow

git add 新的文件
git add -u 更改的文件
git add -a 以上所有
git push 将文件push到github上
git checkout -b branchname 创建分支
git checkout master 回到主版本
在github网站上执行 pull request 将改变merge到其他branch

基本的markdown

哎呀 发现没有字幕也能听懂

R package 从CRAN网站上下载
available.package() 寻找package
install.package(“name”) 安装package
install.package(c(“name1”,”name2”))一次安装多个
或者Studio -》 tools -》 install
library(name)导入 R
search() 查看所有package

data science希望解决哪些问题

descriptive analyses 主要是画图？
exploratory analysis 数据之间的关系？e.g. sloan digital sky survey
inference analysis从部分样本中总结一个对全体样本的结论？
prediction analysis 比较富有挑战性
causal analysis 因果分析
mechanistic analysis