Menu
I have a matrix containing integers and a data frame with several columns.
Matrix:
[,1] [,2] [,3] [,4] [,5] [,6]
[1,] 1 4 6 1 NA NA
[2,] 2 3 NA NA NA NA
[3,] 3 4 5 6 2 1
[4,] 6 6 2 3 3 NA
[5,] 1 2 1 4 5 6
[6,] 4 NA NA NA NA NA
Data frame:
V1 V2 V3
1 "5P" "Fox" "28639"
2 "5P" "Horse" "33844"
3 "5P" "Cat" "Bes86"
4 "5P" "Seal" "Bes259"
5 "5P" "Snake" "Bes260"
6 "5P" "Platypus" "NSA8631"
The actual data frame is much larger than this (10000+ rows).
What I want is to replace the numbers in the matrix with the corresponding row of V2 in my data frame. So that all "1" entries end up as "Fox", "2" as "Horse" and so on.
[,1] [,2] [,3] [,4] [,5] [,6]
[1,] Fox Seal Platypus Fox NA NA
[2,] Horse Cat NA NA NA NA
[3,] Cat Seal Snake Platypus Horse Fox
[4,] Platypus Platypus Horse Cat Cat NA
[5,] Fox Horse Fox Seal Snake Platypus
[6,] Seal NA NA NA NA NA
Thanks for any assistance!
542
You can specify the value to replace as c , M(1,3) or M(3,1) . To replace a particular element of a matrix with a new value while keeping all other elements unchanged, use the assignment operation. For example, M(1,1) = 2 replaces only the first element of the matrix M with the value 2.
Description. newStr = replace( str , old , new ) replaces all occurrences of the substring old with new . If old contains multiple substrings, then new either must be the same size as old , or must be a single substring.
You might have heard that matrix can contain only numerical values but it is also possible to create a matrix with string values, and of course calculations using these types of matrices would not be possible.
Convert a string to int : int() int() converts a string to an integer int . A string containing . or , raises an error. A comma-separated string can be converted by deleting , with replace() (replace with the empty string '' ).
This seems to do the trick:
#you convert the matrix to vector
#use it to index df2$V2
#and then reconstruct the matrix
matrix(df2$V2[as.vector(mat)], ncol=6)
#Or actually even better as @PierreLafortune messaged me
#you don't even need as.vector as this occurs automatically
matrix(df2$V2[mat], ncol=ncol(mat)) #result is the same
Data:
mat <- as.matrix(read.table(header=T,text=' [,1] [,2] [,3] [,4] [,5] [,6]
[1,] 1 4 6 1 NA NA
[2,] 2 3 NA NA NA NA
[3,] 3 4 5 6 2 1
[4,] 6 6 2 3 3 NA
[5,] 1 2 1 4 5 6
[6,] 4 NA NA NA NA NA'))
df2 <- read.table(text='V1 V2 V3
1 "5P" "Fox" "28639"
2 "5P" "Horse" "33844"
3 "5P" "Cat" "Bes86"
4 "5P" "Seal" "Bes259"
5 "5P" "Snake" "Bes260"
6 "5P" "Platypus" "NSA8631" ')
Output:
[,1] [,2] [,3] [,4] [,5] [,6]
[1,] "Fox" "Seal" "Platypus" "Fox" NA NA
[2,] "Horse" "Cat" NA NA NA NA
[3,] "Cat" "Seal" "Snake" "Platypus" "Horse" "Fox"
[4,] "Platypus" "Platypus" "Horse" "Cat" "Cat" NA
[5,] "Fox" "Horse" "Fox" "Seal" "Snake" "Platypus"
[6,] "Seal" NA NA NA NA NA
You can also use lookup from qdapTools:
library(qdapTools)
matrix(lookup(c(mat), data.frame(1:nrow(df2),df2$V2)), ncol=ncol(mat))
# [,1] [,2] [,3] [,4] [,5] [,6]
#[1,] "Fox" "Seal" "Platypus" "Fox" NA NA
#[2,] "Horse" "Cat" NA NA NA NA
#[3,] "Cat" "Seal" "Snake" "Platypus" "Horse" "Fox"
#[4,] "Platypus" "Platypus" "Horse" "Cat" "Cat" NA
#[5,] "Fox" "Horse" "Fox" "Seal" "Snake" "Platypus"
#[6,] "Seal" NA NA NA NA NA
If you love us? You can donate to us via Paypal or buy me a coffee so we can maintain and grow! Thank you!
Donate Us With
