标签：r语言   数据分析   数据挖掘   机器学习   

1.       timestamp函数输出当前时间

timestamp()

##------ Sun Apr 05 20:54:06 2015 ------##

该函数可以输入当前的系统时间，可用于耗时很长的程序定时输出当前时间，用于判断程序是否正常运行；也可用于调试，判断哪一段代码效率较低。

2.       多个比较的boxplot图

a=c(1,2,3,4,5,2,1,2,4,2,5,6)

b=c("A","A","B","B","C","C","C","C","D","D","D","D")

c=c(1,2,3,4,5,5,2,1,2,4,2,5,6,5,6)

d=c("A","A","A","A","B","B","B","C","C","C","C","C","D","D","D")

boxplot(a~b,boxwex=.24,at=1:4-.2,col="orange",axes=FALSE)

# 定义箱式图的宽度

box();axis(2)

axis(1,at=c(1:4-.2,1:4+.2),rep(c("\u2640","\u2642"),each=4))

boxplot(c~d,boxwex=.24,at=1:4+.2,col="yellow",add=TRUE,axes=FALSE)# add

3.       绘制地图

require(maps)

map(database = "state", fill =FALSE, boundary = TRUE, interior = FALSE, col = "grey60", lwd = 0.6,lty = 1, add = FALSE, resolution = 0)

map(database = "state", fill =FALSE, boundary = FALSE, interior = TRUE, col = "grey70", lwd = 0.5,lty = "31", add = TRUE, resolution = 0)

par(cex.axis = 0.7)

map.axes()

x1 = c(-109, -108, -103, -99, -97, -99,-107)

y1 = c(39, 43, 48, 43, 39, 33, 37)

xspline(x = x1, y = y1, shape = 0.9, open =FALSE, border = NA, col = "#00FF0037")

x2 = c(-107, -107, -103, -101, -99, -101,-105)

y2 = c(39, 41, 43, 41, 39, 37, 38)

xspline(x = x2, y = y2, shape = 0.9, open =FALSE, border = NA, col = "#FF000073")

points(-103, 39, pch = 19)

require(fields)

data(ozone2)

x<-ozone2$lon.lat

y<- ozone2$y[16,] # June 18, 1987

fit<- Tps(x,y)

surface(fit)

US(add=TRUE, col="magenta",lwd=2)

title("Daily max 8 hour ozone inPPB,  June 18th, 1987")

4.       绘制双重坐标轴

绘制双重坐标轴的图形不是一个好的图形，会给读者带来很大的迷惑。

plot(1, main = ‘如何折磨读者的眼睛‘)

axis(2, runif(7, .6, 1.4), line=1)

axis(4, runif(3, .6, 1.4), line=1)

5.       绘制密度函数和分布函数曲线

x=seq(-5,5,0.1)

plot(x,dnorm(x),type="l")   #密度函数曲线

plot(x,pnorm(x),type="l")   #分布函数曲线

6.       三维图中对Z坐标的要求

z是matrix！（不是函数）你要理解三维图形和二维图形是不一样的，并不是说有三个向量x、y、z就能画出3D图来的（除非是只画点），3D图形通常都是一个曲面，因此需要一个grid形式的数据，也就是每一对(x, y)对应的z值。如：

persp(x=1:10, y=1:10,z=matrix(rnorm(10*10),10,10))

7.      判断数据框是否相同

df1 == df2

这样比较可以判断数据框中每一对应元素是不是相等。

8.       用exists函数判断变量是否已经存在

if(!exists("dataPath")){

dataPath<-choose.dir(default ="", caption = "Please select a data source folder")

}

9.      为散点图的点做标注

x=c(1,1,2,2,1,1,2)

y=c(1,2,2,5,5,3,3)

lbls <- c(

"list(x1,y1^(1))","list(x2,y2^(2))","list(x3,y3^(3))",

"list(x4,y4^(4))","list(x5,y5^(5))","list(x6,y6^(6))",

"list(x7,y7^(7))"

)

lbls <- paste("(",lbls,")", "")

{plot(x,y,main="X vsY",xlab="X", ylab="Y",xlim=c(.9,2.1),ylim=c(.9,5))

lines(x,y,col="black",lty="dotted")

text(x,y-.1,labels=parse(text=lbls))}

10. density函数的解释

x=rnorm(1000)   # x是你希望对什么数据作核密度估计，density()中n的默认值为512

n=1000          # n是你希望对多少个点作核密度估计，x不太长时可指定n=length(x)

a=density(x,n)  # a是density()给出的全部计算结果

a[1:2]          # 你想要的密度估计曲线上所有点的坐标，其中

a$y             # density()给出的核密度估计值，即图中的纵坐标

a$x             # 与a$y对应的横坐标

density.default # density()函数代码

11. 利用getAnywhere函数查找R对象

有些函数的源代码是隐藏状态，可以尝试用下列方法获取：

getAnywhere(princomp.default)

12. 绘制极坐标图形

library(plotrix)

t<-seq(-pi,pi,0.1)

r<-1-cos(t/2)

radial.plot(r,t,rp.type="p",lwd=3,line.col=2)

13. 利用assign函数对字符串形式的变量名赋值

for(i in 1:6) { #-- Create objects  ‘r.1‘, ‘r.2‘, ... ‘r.6‘ --

   nam <- paste("r", i, sep = ".")

   assign(nam, 1:i)

}

ls(pattern = "^r..$")

##-- Global assignment within a function:

myf <- function(x) {

   innerf <- function(x) assign("Global.res", x^2, envir =.GlobalEnv)

   innerf(x+1)

}

myf(3)

Global.res # 16

a <- 1:4

assign("a[1]", 2)

a[1] == 2          # FALSE

get("a[1]") == 2   # TRUE

14. 负数开方运算

可转化为复数的运算

(-1+0i)^(1/3)

[1] 0.5+0.8660254i

15. R语言中表示“空集”

以下是不同数据类型的“空集”

list()

c()

data.frame()

factor()

R语言基础编程技巧汇编 - 17

标签：r语言   数据分析   数据挖掘   机器学习   

原文地址：http://blog.csdn.net/liu7788414/article/details/44891279