I have a data frame with two columns. The first one "V1" indicates the objects on which the different items of the second column "V2" are found, e.g.:
V1 <- c("A", "A", "A", "A", "B", "B", "B", "C", "C", "C", "C")
V2 <- c("a","b","c","d","a","c","d","a","b","d","e")
df <- data.frame(V1, V2)
"A" for example contains "a", "b", "c", and "d". What I am looking for is a three dimensional array with dimensions of length(unique(V2)) (and the names "a" to "e" as dimnames).
For each unique value of V1 I want all possible combinations of three V2 items (e.g. for "A" it would be c("a", "b", "c"), c("a", "b", "d", and c("b", "c", "d").
Each of these "three-item-co-occurrences" should be regarded as a coordinate in the three-dimensional array and therefore be added to the frequency count which the values in the array should display. The outcome should be the following array
ar <- array(data = c(0,0,0,0,0,0,0,1,2,1,0,1,0,2,0,0,2,2,0,1,0,1,0,1,0,
0,0,1,2,1,0,0,0,0,0,1,0,0,1,0,2,0,1,0,1,1,0,0,1,0,
0,1,0,2,0,1,0,0,1,0,0,0,0,0,0,2,1,0,0,0,0,0,0,0,0,
0,2,2,0,1,2,0,1,0,1,2,1,0,0,0,0,0,0,0,0,1,1,0,0,0,
0,1,0,1,0,1,0,0,1,0,0,0,0,0,0,1,1,0,0,0,0,0,0,0,0),
dim = c(5, 5, 5),
dimnames = list(c("a", "b", "c", "d", "e"),
c("a", "b", "c", "d", "e"),
c("a", "b", "c", "d", "e")))
I was wondering about the 3D symmetry of your result. It took me a while to understand that you want to have all permutations of all combinations.
library(gtools) #for the permutations
foo <- function(x) {
#all combinations:
combs <- combn(x, 3, simplify = FALSE)
#all permutations for each of the combinations:
combs <- do.call(rbind, lapply(combs, permutations, n = 3, r = 3))
#tabulate:
do.call(table, lapply(asplit(combs, 2), factor, levels = letters[1:5]))
}
#apply grouped by V1, then sum the results
res <- Reduce("+", tapply(df$V2, df$V1, foo))
#check
all((res - ar)^2 == 0)
#[1] TRUE
I used to use the crossjoin CJ() to retain the pairwise count of all combinations of two different V2 items
res <- setDT(df)[,CJ(unique(V2), unique(V2)), V1][V1!=V2,
.N, .(V1,V2)][order(V1,V2)]
This code creates a data frame res with three columns. V1 and V2 contain the respective items of V2 from the original data frame df and N contains the count (how many times V1 and V2 appear with the same value of V1 (from the original data frame df).
Now, I found that I could perform this crossjoin with three 'dimensions' as well by just adding another unique(V2) and adapting the rest of the code accordingly.
The result is a data frame with four columns. V1, V2, and V3 indicate the original V2 items and N again shows the number of mutual appearances with the same original V1 objects.
res <- setDT(df)[,CJ(unique(V2), unique(V2), unique(V2)), V1][V1!=V2 & V1 != V3 & V2 != V3,
.N, .(V1,V2,V3)][order(V1,V2,V3)]
The advantage of this code is that all empty combinations (those which do not appear at all) are not considered. It worked with 1,000,000 unique values in V1 and over 600 unique items in V2, which would have otherwise caused an extremely large array of 600 x 600 x 600
I'm struggling with having the subset() function use a range (i.e. 4:7) that is being called as a character from a variable.
Is there a way for me to coerce the input, which is the variable DayVar and has different days I want the function to subset, to be numeric while avoiding the following issues:
1.) keeping the 4:7 as such instead of as 4, 5, 6, 7, and
2.) converting the character "1:4" into numeric format that the subset evaluation can use as though it were 1:4.
Here is a sample data frame:
DayVar = c("1", "2", "3", "4:7")
a <- c("a", "b", "c", "d", "e", "f", "g", "h", "i", "j")
b <- c(61:70)
Day <- c(1:10)
df <- data.frame("a" = a, "b" = b, "Day" = Day)
Subset <- list()
for(i in 1:length(DayVar)){
Subset[[i]] = subset(df, Day %in% DayVar[i])
}
As thelatemail suggested the list works but you have to change the DayVar quotes to get the list index:
DayVar <- list(1,2,3,4:7)
Subset <- list()
for(i in 1:length(DayVar)){
Subset[[i]] = subset(df, Day %in% DayVar[[i]])
}
Suppose I have the following data frame:
foo <- data.frame(a=letters,b=seq(1,26),
n1=rnorm(26),n2=rnorm(26),
u1=runif(26),u2=runif(26))
I want to append columns u1 and u2 to columns n1 and n2. For now, I found the following way:
df1 <- foo[,c("a","b","n1","n2")]
df2 <- foo[,c("a","b","u1","u2")]
names(df2) <- names(df1)
bar <- rbind(df1,df2)
That does the trick. However, it seems a little bit involved. Am I too picky? Or is there a faster/simpler way to do this in R?
Here is one way using full_join() from dplyr:
library(dplyr)
full_join(df1, df2, by = c("a", "b", "n1" = "u1", "n2" = "u2"))
From the documentation:
full_join
return all rows and all columns from both x and y. Where
there are not matching values, returns NA for the one missing.
by
a character vector of variables to join by. If NULL, the default,
join will do a natural join, using all variables with common names
across the two tables. A message lists the variables so that you can
check they're right.
To join by different variables on x and y use a named vector. For
example, by = c("a" = "b") will match x.a to y.b.
Use Map() to concatenate the columns, and cbind() with recycling to arrive at the final data frame.
cbind(foo[1:2], Map(c, foo[3:4], foo[5:6]))
Substitute numerical indexes with column names, if desired.
cbind(foo[c("a", "b")], Map(c, foo[c("n1", "n2")], foo[c("u1", "u2")]))
Short-hand:
rbind(foo[1:4], setNames(foo[c(1, 2, 5, 6)], names(foo[1:4])))
Long-winded:
rbind(foo[c("a", "b", "n1", "n2")], setNames(foo[c("a", "b", "u1", "u2")], c("a", "b", "n1", "n2")))
Long-winded (more DRY):
nms <- c("a", "b", "n1", "n2")
rbind(foo[nms], setNames(foo[c("a", "b", "u1", "u2")], nms))
I'm trying to reorder the rows of a data frame by two factors. For the first factor i'm happy with the default ordering. For the second factor i'd like to impose my own custom order to the rows. Here's some dummy data:
dat <- data.frame(apple=rep(LETTERS[1:10], 3),
orange=c(rep("agg", 10), rep("org", 10), rep("fut", 10)),
pear=rnorm(30, 10),
grape=rnorm(30, 10))
I'd like to order "apple" in a specific way:
appleOrdered <- c("E", "D", "J", "A", "F", "G", "I", "B", "H", "C")
I've tried this:
dat <- dat[with(dat, order(orange, rep(appleOrdered, 3))), ]
But it seems to put "apple" into a random order. Any suggestions? Thanks.
Reordering the factor levels:
dat[with(dat, order(orange, as.integer(factor(apple, appleOrdered)))), ]
Try using a factor with the levels in the desired order and the arrange function from plyr:
dat$apple <- factor(dat$apple,levels=appleOrdered)
arrange(dat,orange,apple)