R fill matrix based on element names from data frame

Question

If I have data that may be modelled as so:

a1 <- c("bob","bill",0.2)
a2 <- c("bob", "bert", 0.1)
a3 <- c("bill", "bert", 0.1)

my.df <- as.data.frame(rbind(a1, a2, a3), stringsAsFactors = FALSE)

colnames(my.df) <- c("name_1", "name_2", "value")
row.names(my.df) <- NULL

my.names <- unique(as.character(c(my.df$name_1, my.df$name_2)))

my.matrix <- matrix(0, nrow = length(my.names), ncol = length(my.names))

row.names(my.matrix) <- my.names

colnames(my.matrix) <- my.names

How may I fill my.matrix using the values from my.df. The first two columns in my.df describe the coordinates of my.matrix to fill?

bgoldst · Accepted Answer · 2016-06-20 20:24:43Z

4

You can coerce the first two columns of my.df into an index matrix, and index-assign my.matrix with the remaining column of my.df as the RHS:

my.matrix[as.matrix(my.df[c('name_1','name_2')])] <- as.double(my.df$value);
my.matrix;
##      bob bill bert
## bob    0  0.2  0.1
## bill   0  0.0  0.1
## bert   0  0.0  0.0

I also coerced the RHS to a double, since I assumed you want to preserve the type of my.matrix.

answered Jun 20, 2016 at 20:24

bgoldst

35.6k6 gold badges44 silver badges64 bronze badges

Sign up to request clarification or add additional context in comments.

2 Comments

mattbawn Over a year ago

Great thanks! This worked best with the real data and I was able to easily adapt it to fill the other half of the matrix.

Sss Over a year ago

I have a follow up question, is it possible when the data is missing to set the cell as empty instead of 0?

toni057 · Accepted Answer · 2016-06-20 20:31:25Z

2

An easy solution using a for loop:

my.df$value <- as.numeric(my.df$value)
for (i in 1:nrow(my.df)) {
   my.matrix[rownames(my.matrix) == my.df$name_1[i], 
             colnames(my.matrix) == my.df$name_2[i]] <- my.df$value[i]
}

answered Jun 20, 2016 at 20:31

toni057

6141 gold badge5 silver badges11 bronze badges

Collectives™ on Stack Overflow

R fill matrix based on element names from data frame

2 Answers 2

2 Comments

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

2 Comments

Comments

Your Answer

Sign up or log in

Post as a guest

Related