矩阵
矩阵（matrix）是一种特殊的向量，包含两个附加的属性：行数和列数，R生存矩阵按列存储
注意：向量不能看成只有一行或者一列的矩阵，二者不能混为一谈
矩阵的创建
- 利用matrix()函数 例如 matrix（data = NA，nrow = 1,ncol = 1,byrow = FALSE,dimnames = NULL）;
data表示要处理的数据（向量）
nrow表示行
ncol表示列
byrow表示是否按行排列，默认是列排
dimnames表示行与列的名字，默认是没有的，要使用列表设置
代码语言：R复制> mydata <- matrix(c(1,2,3,4,5,6),nrow = 2,ncol = 3)#两行三列
> mydata
[,1] [,2] [,3]
[1,] 1 3 5
[2,] 2 4 6
> mydata1 <- matrix(c(1,2,3,4,5,6),nrow = 2,ncol = 3,byrow = T)
> mydata1
[,1] [,2] [,3]
[1,] 1 2 3
[2,] 4 5 6
> mydata3 <- matrix(c(1,2,3,4,5,6),nrow = 2,ncol = 3,dimnames = list(c("r1","r2"),c("c1","c2","c3")))
> mydata3
c1 c2 c3
r1 1 3 5
r2 2 4 62.利用函数将向量接起来
rbind()函数：按行接起来
cbind()函数：按列接起来
代码语言：R复制> mydata5 <- rbind(c(1,2,3),c(4,5,6))
> mydata6 <- cbind(c(1,2,3),c(4,5,6))
> mydata5
[,1] [,2] [,3]
[1,] 1 2 3
[2,] 4 5 6
> mydata6
[,1] [,2]
[1,] 1 4
[2,] 2 5
[3,] 3 6
> mydata7 <- rbind(c(1,2,3,4,5),c(4,5,6))#出现警告，循环补齐
Warning message:
In rbind(c(1, 2, 3, 4, 5), c(4, 5, 6)) :
number of columns of result is not a multiple of vector length (arg 2)
> mydata7
[,1] [,2] [,3] [,4] [,5]
[1,] 1 2 3 4 5
[2,] 4 5 6 4 5
> mydata8 <- cbind(c(1,2,3),c(4,5,6,7,8))
Warning message:
In cbind(c(1, 2, 3), c(4, 5, 6, 7, 8)) :
number of rows of result is not a multiple of vector length (arg 1)
> mydata8
[,1] [,2]
[1,] 1 4
[2,] 2 5
[3,] 3 6
[4,] 1 7
[5,] 2 83.强制转换为矩阵
利用as.matrix()函数
代码语言：R复制> a <- c(1,2,3,4,5)
> a1 <- as.matrix(a)
> a1
[,1]
[1,] 1
[2,] 2
[3,] 3
[4,] 4
[5,] 54.建议一个空矩阵
直接写行列大小
代码语言：R复制> b <- matrix(nrow = 3,ncol = 2)
> b
[,1] [,2]
[1,] NA NA
[2,] NA NA
[3,] NA NA注意：当向量中含有不同类型的数据时，会改变元素类型后转变为矩阵
代码语言：R复制> mydata10 <- matrix(c(2,3,FALSE,TRUE),2,2) #向量中含有逻辑与数值型，逻辑转为数值，2行2列排序
> mydata10
[,1] [,2]
[1,] 2 0
[2,] 3 1
> #当向量含有逻辑、字符、数值型，全部转换为字符
> mydata11 <- matrix(c(2,3,5,7,"ABC",FALSE,TRUE,3,9,3,4,5),4,3)
> mydata11
[,1] [,2] [,3]
[1,] "2" "ABC" "9"
[2,] "3" "FALSE" "3"
[3,] "5" "TRUE" "4"
[4,] "7" "3" "5" 矩阵的索引
使用下标和中括号来选择矩阵中的行或列或元素，矩阵名 中括号，中括号里面写需要的行与列
例如，设置一个矩阵，名字为mymatrix，5行5列
代码语言：R复制> mymatrix <- matrix(c(1:25),5,5)
> mymatrix
[,1] [,2] [,3] [,4] [,5]
[1,] 1 6 11 16 21
[2,] 2 7 12 17 22
[3,] 3 8 13 18 23
[4,] 4 9 14 19 24
[5,] 5 10 15 20 25
> #矩阵每一行前或列首，都有一个中括号，可以理解为这一行或列标
> #mymatrix[i,]，返回矩阵mymatrix中第i行元素
> mymatrix[3,]
[1] 3 8 13 18 23
> #mymatrix[,j]，返回矩阵mymatrix中第j列元素
> mymatrix[,5]
[1] 21 22 23 24 25
> mymatrix[3,4]
[1] 18
> mymatrix[2,-5]#返回第i行，但是排除第j列
[1] 2 7 12 17
> mymatrix[c(1,3),c(3,5)]#返回第1和3行，第3和5列
[,1] [,2]
[1,] 11 21
[2,] 13 23
> mymatrix[-c(1,3),c(3,5)]#返回第3和5列，但是排除第1行和第3行
[,1] [,2]
[1,] 12 22
[2,] 14 24
[3,] 15 25
#当矩阵有行列名时，可以通过行与列名进行提取元素
> mymatrix <- matrix(c(1:25),5,5,dimnames = list(c("r1","r2","r3","r4","r5"),c("c1","c2","c3","c4","c5")))
> mymatrix
c1 c2 c3 c4 c5
r1 1 6 11 16 21
r2 2 7 12 17 22
r3 3 8 13 18 23
r4 4 9 14 19 24
r5 5 10 15 20 25
> mymatrix["r4","c1"]
[1] 4
> mymatrix[c("r3","r5"),c("c1","c3")]
c1 c3
r3 3 13
r5 5 15drop处理意外降维
上面矩阵的引用1~5条，从一个矩阵中提取的结果都为向量，如果说，我取一行或者一列返回结果要求是矩阵，虽然说可以先去后变为矩阵 但是相对麻烦，我们可以加入drop防止降维
代码语言：R复制> mymatrix1 <- matrix(c(1:36),6,6,dimnames = list(c("r1","r2","r3","r4","r5","r6"),c("c1","c2","c3","c4","c5","c6")))
> mymatrix1
c1 c2 c3 c4 c5 c6
r1 1 7 13 19 25 31
r2 2 8 14 20 26 32
r3 3 9 15 21 27 33
r4 4 10 16 22 28 34
r5 5 11 17 23 29 35
r6 6 12 18 24 30 36
> mymatrix1[2,]#直接取会降维为向量
c1 c2 c3 c4 c5 c6
2 8 14 20 26 32
> mymatrix1[6,,drop=F]#如果想取出是矩阵，则加入drop=F
c1 c2 c3 c4 c5 c6
r6 6 12 18 24 30 36矩阵相关操作及相关函数
1.转置，函数t()
代码语言：R复制> #首先创建一个矩阵
> mymatrix <- matrix(c(1:36),6,6,dimnames = list(c("r1","r2","r3","r4","r5","r6"),c("c1","c2","c3","c4","c5","c6")))
> mymatrix
c1 c2 c3 c4 c5 c6
r1 1 7 13 19 25 31
r2 2 8 14 20 26 32
r3 3 9 15 21 27 33
r4 4 10 16 22 28 34
r5 5 11 17 23 29 35
r6 6 12 18 24 30 36
> #通过t()函数进行转置，返回结果如下
> mymatrix3 <- t(mymatrix)
> mymatrix3
r1 r2 r3 r4 r5 r6
c1 1 2 3 4 5 6
c2 7 8 9 10 11 12
c3 13 14 15 16 17 18
c4 19 20 21 22 23 24
c5 25 26 27 28 29 30
c6 31 32 33 34 35 362.横向或者纵向合并矩阵
cbind()函数横向合并矩阵，rbind()函数纵向合并矩阵
代码语言：R复制> M <- matrix(c(1:36),6,6,dimnames = list(c("r1","r2","r3","r4","r5","r6"),c("c1","c2","c3","c4","c5","c6")))
> M
c1 c2 c3 c4 c5 c6
r1 1 7 13 19 25 31
r2 2 8 14 20 26 32
r3 3 9 15 21 27 33
r4 4 10 16 22 28 34
r5 5 11 17 23 29 35
r6 6 12 18 24 30 36
> N <- matrix(c(37:72),6,6,dimnames = list(c("a1","a2","a3","a4","a5","a6"),c("b1","b2","b3","b4","b5","b6")))
> N
b1 b2 b3 b4 b5 b6
a1 37 43 49 55 61 67
a2 38 44 50 56 62 68
a3 39 45 51 57 63 69
a4 40 46 52 58 64 70
a5 41 47 53 59 65 71
a6 42 48 54 60 66 72
> A <- cbind(M,N)
> B <- rbind(M,N)
> A
c1 c2 c3 c4 c5 c6 b1 b2 b3 b4 b5 b6
r1 1 7 13 19 25 31 37 43 49 55 61 67
r2 2 8 14 20 26 32 38 44 50 56 62 68
r3 3 9 15 21 27 33 39 45 51 57 63 69
r4 4 10 16 22 28 34 40 46 52 58 64 70
r5 5 11 17 23 29 35 41 47 53 59 65 71
r6 6 12 18 24 30 36 42 48 54 60 66 72
> B
c1 c2 c3 c4 c5 c6
r1 1 7 13 19 25 31
r2 2 8 14 20 26 32
r3 3 9 15 21 27 33
r4 4 10 16 22 28 34
r5 5 11 17 23 29 35
r6 6 12 18 24 30 36
a1 37 43 49 55 61 67
a2 38 44 50 56 62 68
a3 39 45 51 57 63 69
a4 40 46 52 58 64 70
a5 41 47 53 59 65 71
a6 42 48 54 60 66 723.对各列或行求和或求均值
colSums()对各列求和，rowMeans()对各行求均值
代码语言：R复制> D <- colSums(M)
> E <- rowMeans(N)
> D
c1 c2 c3 c4 c5 c6
21 57 93 129 165 201
> E
a1 a2 a3 a4 a5 a6
52 53 54 55 56 574.计算行列式
det（）函数，计算行列式
代码语言：R复制> U1 <- matrix(c(1:4),2,2,dimnames = list(c("r1","r2"),c("c1","c2")))
> U2 <- det(U1)
> U2
[1] -25.矩阵相乘
使用%*%进行矩阵相乘
代码语言：R复制> U3 <- matrix(c(5:8),2,2)
> K <- U1%*%U3
> K
[,1] [,2]
r1 23 31
r2 34 466.设置行列名
设置行列名有三种以上的方式，介绍三种常用的
代码语言：R复制> #使用dimnames，以列表的形式设置行列名
> fa <- matrix(c(1:4),2,2,dimnames = list(c("第一行","第二行"),c("第一列","第二列")))
> fa
第一列 第二列
第一行 1 3
第二行 2 4
> #在外部，使用dimnames()函数设置行列名
> f1 <- matrix(c(6:9), nrow = 2, ncol = 2)
> a <- c("第一行","第二行")
> b <- c("第一列","第二列")
> dimnames(f1) <- list(a,b)
> f1
第一列 第二列
第一行 6 8
第二行 7 9
> #使用函数rownames()设置行名，使用colnames()设置列名
> f2 <- matrix(c(1,3,5,7),2,2)
> rownames(f2) <- c("A1","A2")
> colnames(f2) <- c("B1","B2")
> f2
B1 B2
A1 1 5
A2 3 7apply()函数
apply(m,dimcode,fargs)，允许用户在各行各列调用函数
m 矩阵，dimcode，1代表行，2代表列，f函数，frags可选参数，函数大于一个参考时使用“，”隔开
代码语言：R复制> mymatrix <- matrix(c(1:25),5,5)
> mymatrix
[,1] [,2] [,3] [,4] [,5]
[1,] 1 6 11 16 21
[2,] 2 7 12 17 22
[3,] 3 8 13 18 23
[4,] 4 9 14 19 24
[5,] 5 10 15 20 25
> mymatrix_rowmean <- apply(mymatrix,1,mean) #对行求均值
> mymatrix_rowmean
[1] 11 12 13 14 15
> mymatrix_colmean <- apply(mymatrix,2,mean) #对列求均值
> mymatrix_colmean
[1] 3 8 13 18 23
> mymatrix_rowmean1 <- matrix(mymatrix_rowmean,length(mymatrix_rowmean) ,1)#转换为一列的矩阵
> mymatrix_rowmean1
[,1]
[1,] 11
[2,] 12
[3,] 13
[4,] 14
[5,] 15
> mymatrix_colmean1 <- matrix(mymatrix_colmean,1,length(mymatrix_colmean))#转换为一行的矩阵
> mymatrix_colmean1
[,1] [,2] [,3] [,4] [,5]
[1,] 3 8 13 18 23
> mymatrix_colmean[6] <- NA #矩阵右下角有个空，使用NA值填一下，方便连接
> A <- cbind(mymatrix,mymatrix_rowmean1) #按照列把两个矩阵连接起来
> B <- rbind(A,mymatrix_colmean1) #按照行把两个矩阵连接起来
Error in rbind(A, mymatrix_colmean1) : 矩阵的列数必需相符(见arg2)
> rm(list = ls())
> mymatrix <- matrix(c(1:25),5,5)
> mymatrix
[,1] [,2] [,3] [,4] [,5]
[1,] 1 6 11 16 21
[2,] 2 7 12 17 22
[3,] 3 8 13 18 23
[4,] 4 9 14 19 24
[5,] 5 10 15 20 25
> mymatrix_rowmean <- apply(mymatrix,1,mean) #对行求均值
> mymatrix_rowmean
[1] 11 12 13 14 15
> mymatrix_colmean <- apply(mymatrix,2,mean) #对列求均值
> mymatrix_colmean
[1] 3 8 13 18 23
> mymatrix_rowmean1 <- matrix(mymatrix_rowmean,length(mymatrix_rowmean) ,1)#转换为一列的矩阵
> mymatrix_rowmean1
[,1]
[1,] 11
[2,] 12
[3,] 13
[4,] 14
[5,] 15
> mymatrix_colmean1 <- matrix(mymatrix_colmean,1,length(mymatrix_colmean))#转换为一行的矩阵
> mymatrix_colmean1
[,1] [,2] [,3] [,4] [,5]
[1,] 3 8 13 18 23
> mymatrix_colmean1[6] <- NA #矩阵右下角有个空，使用NA值填一下，方便连接
> A <- cbind(mymatrix,mymatrix_rowmean1) #按照列把两个矩阵连接起来
> B <- rbind(A,mymatrix_colmean1) #按照行把两个矩阵连接起来
> a <- c("r1","r2","r3","r4","r5","列均值")
> b <- c("a1","a3","a3","a4","a5","行均值",)
Error in c("a1", "a3", "a3", "a4", "a5", "行均值", ) : 7元素是空的
> rm(list = ls())
> mymatrix <- matrix(c(1:25),5,5)
> mymatrix
[,1] [,2] [,3] [,4] [,5]
[1,] 1 6 11 16 21
[2,] 2 7 12 17 22
[3,] 3 8 13 18 23
[4,] 4 9 14 19 24
[5,] 5 10 15 20 25
> mymatrix_rowmean <- apply(mymatrix,1,mean) #对行求均值
> mymatrix_rowmean
[1] 11 12 13 14 15
> mymatrix_colmean <- apply(mymatrix,2,mean) #对列求均值
> mymatrix_colmean
[1] 3 8 13 18 23
> mymatrix_rowmean1 <- matrix(mymatrix_rowmean,length(mymatrix_rowmean) ,1)#转换为一列的矩阵
> mymatrix_rowmean1
[,1]
[1,] 11
[2,] 12
[3,] 13
[4,] 14
[5,] 15
> mymatrix_colmean1 <- matrix(mymatrix_colmean,1,length(mymatrix_colmean))#转换为一行的矩阵
> mymatrix_colmean1
[,1] [,2] [,3] [,4] [,5]
[1,] 3 8 13 18 23
> mymatrix_colmean1[6] <- NA #矩阵右下角有个空，使用NA值填一下，方便连接
> A <- cbind(mymatrix,mymatrix_rowmean1) #按照列把两个矩阵连接起来
> B <- rbind(A,mymatrix_colmean1) #按照行把两个矩阵连接起来
> a <- c("r1","r2","r3","r4","r5","列均值")
> b <- c("a1","a2","a3","a4","a5","行均值")
> dimnames(B) <- list(a,b)
> B
a1 a2 a3 a4 a5 行均值
r1 1 6 11 16 21 11
r2 2 7 12 17 22 12
r3 3 8 13 18 23 13
r4 4 9 14 19 24 14
r5 5 10 15 20 25 15
列均值 3 8 13 18 23 NA数据结构的查看
可以使用class（）函数查看数据结构，如何判断对象是否是矩阵呢？
1.使用attributes（）函数，若是矩阵返回一个维度，若不是，返回NULL值
代码语言：R复制> fa <- matrix(c(1:4),2,2) #返回维度，2行2列
> attributes(fa)
$dim
[1] 2 2
> attributes(fa[2,])#向量无维度，返回NULL值
NULL
> is.matrix(fa) #判断fa是不是一个矩阵，若是返回TRUE，不是返回NULL
[1] TRUE
> is.vector(fa[2,]) #判断fa是不是一个向量，若是返回TRUE，不是返回NULL
[1] TRUE
