标签：workplace 应该 数学 com 分析 sub 线性回归 个数 apply
*一些操作：
运行R脚本：source (“Rscript”)
当前目录下的文件：dir()
*画图
########R语言画图
###RMSD
getwd()
file <- read.csv("backbone_rmsd.csv",sep=‘‘,header = FALSE)
options(max.print = 5000)
file
y <- file[,1]
x <- c(1:length(y))
#plot(x,y,type="l",main="RMSD",xlab = "time",ylab = "rmsd",col = ‘red‘,ylim=c(1,4))
file2 <- read.table("bcl_rmsd.dat",sep=‘‘,header = FALSE)
file2
y2 <- file2[,1]
x2 <- c(1:length(y))
file3 <- read.table("rg1_rmsd.dat",sep=‘‘,header = FALSE)
file3
y3 <- file3[,1]
x3 <- c(1:length(y))
plot(y,type="l",main="RMSD",xlab = "time",ylab = "rmsd",col = ‘red‘,ylim=c(1,4))
lines(y2,type="l",main="RMSD",xlab = "time",ylab = "rmsd",col = ‘green‘,ylim=c(1,4))
lines(y3,type="l",main="RMSD",xlab = "time",ylab = "rmsd",col = ‘black‘,ylim=c(1,4))
###测试数据
x <- 1:10
y <- 31:40
m<- matrix(1:100,10,10)
z<-5:14
plot(x)
bg="red"
par(bg="gray",lty=2,font=4,ps=18)
plot(x,y,axes = TRUE,sub = "sd",main = "dfsdf",type="l")
text(2,32,"text")
pie(x)
boxplot(x,y,z)
hist(x)
barplot(x)
points(1,2)
lines(c(1:5))
*传课初学：
getwd()
a <- 4
print (a)
class(a)
#####matrix矩阵创建
mat <- matrix(c(1:12),nrow=3,ncol=4)
mat
dim(mat)
b = mat*3
b
mat2 <- matrix(c(1:16),nrow=4)
mat2
c = mat %*% mat2
c
colnames(mat) <- c(‘语文‘,‘数学‘,‘英语‘,‘体育‘)
rownames(mat) <- c(‘小明‘,‘小花‘,‘小刚‘)
mat
#矩阵选择
mat[1,2]
mat[2,]
mat[,3]
mat[c(2,3),c(1,2)]
mat[c(2,3),c(3,4)]
#矩阵筛选
mat[mat[,2] >= 5,]
which(mat[,4] >= 11)
mat[which(mat[,4] >= 11),]
#矩阵操作apply
mean(mat[,2])
apply(mat,1,mean)
apply(mat,2,mean)
apply(mat,2,sum)
###逻辑运算,&,|,略
a <- c(3,2,2)
all(a == 2)
any(a == 2)
3 %in% a
which(a == 2)
###for循环
a <- c(‘a‘,‘b‘,‘c‘)
for (i in c(1:length(a))) {print (i)}
for (i in c(1:length(a))) {print (a[i])}
for (i in a) {print (i)}
a <- matrix(c(1:16),nrow=4)
print(a)
for (i in c(1:nrow(a))) for (j in c(1:ncol(a))) {
print(a[i,j])
}
###repeat循环
###while循环
s <- 0
i <- 1
while(i <= 100){
s <- s + i
i <- i + 1
}
s
###自定义函数
printtest <- function(x){print(x)}
printtest("hello,world!")
ret <- function(x,y){return(x+y)}
ret(2,4)
###文本读取
#设置当前操作路径
getwd()
setwd("D:\\Rstudio-workplace\\home1")
setwd("D:/Rstudio-workplace/home1")
#！！！处理的文件最后一行应该是空行
dat <- read.table("1.txt",sep=‘ ‘,header = TRUE)
dat
write.table(dat,"11.txt",sep=‘ ‘)
write.table(dat,"111.txt",sep=‘,‘,row.names = FALSE,col.names = FALSE)
#读写csv文件
a = read.csv("1.csv",header = FALSE)
dim(a)
###画图
#条形图
x <- c(1:3)
y <- x+2.5
plot(x,y)
plot(x,y,type = "l")
#example
dim(cars)
plot(cars,type = "o",main = ‘测试图‘,xlab="速度",ylab = "dist")
#条形图
barplot(c(88,79,99),col=rainbow(3),names.arg = c(‘小明‘,"小红","小刚"),ylim=c(0,100),legend.text = c(‘小明‘,"小红","小刚"))
#频率分布直方图
hist(iris[,1],col = rainbow(8))
#饼图
pie(c(1,2,3),labels= c("中国","美国","日本"),radius = 0.7)
#箱图（盒图）
boxplot(iris[,c(1:4)],notch = TRUE,col = rainbow(4),names=c("a","b","c","d"))
#图形组合
par(mfrow=c(2,2))
plot(c(1:3))
pie(c(1:3))
boxplot(iris[,c(1:4)],notch = TRUE,col = rainbow(4),names=c("a","b","c","d"))
plot(cars,type = "o",main = ‘测试图‘,xlab="速度",ylab = "dist")
par(mfrow=c(1,1))
###线性回归
s1 <- scale(iris[,1])
pw <- scale(iris[,4])
model <- lm(pw~s1)
summary(model)
model <- lm(pw~s1-1)
summary(model)
plot(pw,s1,col=‘green‘)
###主成分分析
sc <- scale(swiss)
#构建模型
pri <- princomp(sc,cor=TRUE)
#确定主成分个数
screeplot(pri,type=‘line‘)
#计算每一个城市在每一个主成分上的得分
pre <- predict(pri)
summary(pre)
pre
#特征值
y <- eigen(cor(sc))
y$values
#计算最终得分
scores <- (y$values[1]*pre[,1]+y$values[2]*pre[,2]+y$values[3]*pre[,3]+y$values[4]*pre[,4])/sum(y$values[1:4])
scores
plot(scores)
标签：workplace 应该 数学 com 分析 sub 线性回归 个数 apply
原文地址：https://www.cnblogs.com/jszd/p/11178721.html
