I have a matrix M consisting of column names from a data frame with one row, such that each column name has just one corresponding value in the data frame. Is there a function to create a new matrix with the corresponding values from the column names in M?
M <- t(data.frame(A=c("label_1","label_2","label_3"),
B=c("label_4","label_5","label_6"),
C=c("label_7","label_8","label_9")))
M
> [,1] [,2] [,3]
A "label_1" "label_2" "label_3"
B "label_4" "label_5" "label_6"
C "label_7" "label_8" "label_9"
df <- data.frame(label_2=5, label_1=0, label_4=7,
label_6=15, label_3=12, label_5=11,
label_9=9, label_8=15, label_7=35)
df
> label_2 label_1 label_4 label_6 label_3 label_5 label_9 label_8 label_7
1 5 0 7 15 12 11 9 15 35
## I want to create a new data.frame with the values from these labels
> [,1] [,2] [,3]
A 0 5 12
B 7 11 15
C 35 15 9
One possible way I'm aware of is to convert the data frame df to a key-value pair, with k = column names and v = values. I could then retrieve the values using:
apply(M,2,function(x){df[df$k==x,"v"]})
But this seems too overcomplicated for what should be a simple operation...
Additionally, I would prefer not to use any libraries outside of dplyr or tidyr to minimize the dependencies needed in my code.
Updated to an easier code using Onyambu's suggestion:
M <- t(data.frame(A=c("label_1","label_2","label_3"),
B=c("label_4","label_5","label_6"),
C=c("label_7","label_8","label_9")))
df <- data.frame(label_2=5, label_1=0, label_4=7,
label_6=15, label_3=12, label_5=11,
label_9=9, label_8=15, label_7=35)
P<-matrix(df[c(M)],nrow(M))
P
> P
[,1] [,2] [,3]
[1,] 0 5 12
[2,] 7 11 15
[3,] 35 15 9