Have a 1000*16 matrix from a simulation with team names as characters. I want to count number of occurrences per team in all 16 columns.
I know I could do apply(test, 2, table) but that makes the data hard to work with afterward since all teams is not included in every column.
If you have a vector that is all the unique team names you could do something like this. I'm counting occurrences here via column to ensure that not every team (in this case letter) is not included.
set.seed(15)
letter_mat <- matrix(
sample(
LETTERS,
size = 1000*16,
replace = TRUE
),
ncol = 16,
nrow = 1000
)
output <- t(
apply(
letter_mat,
1,
function(x) table(factor(x, levels = LETTERS))
)
)
head(output)
A B C D E F G H I J K L M N O P Q R S T U V W X Y Z
[1,] 1 2 0 1 1 1 1 0 0 0 1 0 0 0 0 1 1 1 1 1 0 1 1 0 0 1
[2,] 0 1 0 2 2 1 0 1 0 0 0 0 0 1 0 0 0 0 0 1 1 1 0 2 2 1
[3,] 1 1 0 0 1 0 1 2 1 0 0 0 0 0 1 0 1 0 1 1 0 0 3 0 1 1
[4,] 0 1 0 0 0 1 0 0 0 2 0 1 0 0 1 1 1 1 2 0 2 3 0 0 0 0
[5,] 2 1 0 0 0 0 0 2 0 2 1 1 1 0 0 2 0 2 1 0 0 1 0 0 0 0
[6,] 0 0 0 0 0 1 3 1 0 0 0 0 1 1 3 0 1 0 0 1 0 0 0 1 0 3
Related
I have created a prediction matrix for large dataset as follows:
library(mice)
dfpredm <- quickpred(df, mincor=.3)
A B C D E F G H I J
A 0 1 1 1 0 1 0 1 1 0
B 1 0 0 0 1 0 1 0 0 1
C 0 0 0 1 1 0 0 0 0 0
D 1 0 1 0 0 1 0 1 0 1
E 0 1 0 1 0 1 1 0 1 0
**F 0 0 1 0 0 0 1 0 0 0**
G 0 1 0 1 0 0 0 0 0 0
H 1 0 1 0 0 1 0 0 0 1
I 0 1 0 1 1 0 1 0 0 0
J 1 0 1 0 0 1 0 1 0 0
I would like to create a subset of the original df on the basis on dfpredm.
More specifically I would like to do the following:
Let's assume that my dependent variable is F.
According to the prediction matrix F is correlated with C and G.
In addition, C and G are best predicted by D,E and B,D respectively.
The idea is now to create a subset of df based on the dependent variable F,for which in the F row the value is 1.
Fpredictors <- df[,(dfpredm["F",]) == 1]
But also do the same for the variables where the rows in F are 1. I am thinking of first getting the column names like this:
Fpredcol <-colnames(dfpredm[,(dfpredm["c241",]) == 1])
And then doing a for loop with these column names?
For the specific example I would like to end up with the subset.
dfsub <- df[,c("F","C","G","B","E","D")]
I would however like to automate this process. Could anyone show me how to do this?
Here is one strategy that seems like it would work for you:
first_preds <- function(dat, predictor) {
cols <- which(dat[predictor, ] == 1)
names(dat)[cols]
}
# wrap first_preds() for getting best and second best predictors
first_and_second_preds <- function(dat, predictor) {
matches <- first_preds(dat, predictor)
matches <- c(matches, unlist(lapply(matches, function(x) first_preds(dat, x))))
c(predictor, matches) %>% unique()
}
dat[first_and_second_preds(dat, "F")] # order is not exactly the same as your output
F C G D E B
A 1 1 0 1 0 1
B 0 0 1 0 1 0
C 0 0 0 1 1 0
D 1 1 0 0 0 0
E 1 0 1 1 0 1
F 0 1 1 0 0 0
G 0 0 0 1 0 1
H 1 1 0 0 0 0
I 0 0 1 1 1 1
J 1 1 0 0 0 0
Not sure if the ordering in the result is important, but you could add the logic if it is.
Using dat from here (a kinder way to share small R data on SO):
dat <- read.table(
text = "A B C D E F G H I J
A 0 1 1 1 0 1 0 1 1 0
B 1 0 0 0 1 0 1 0 0 1
C 0 0 0 1 1 0 0 0 0 0
D 1 0 1 0 0 1 0 1 0 1
E 0 1 0 1 0 1 1 0 1 0
F 0 0 1 0 0 0 1 0 0 0
G 0 1 0 1 0 0 0 0 0 0
H 1 0 1 0 0 1 0 0 0 1
I 0 1 0 1 1 0 1 0 0 0
J 1 0 1 0 0 1 0 1 0 0",
header = TRUE
)
Something a little more general that would let you use self_select predictors directly:
all_preds <- function(dat, predictors) {
unlist(lapply(predictors, function(x) names(dat)[which(dat[x, ] == 1 )]))
}
dat[all_preds(dat, c("A", "B"))]
B C D F H I A E G J
A 1 1 1 1 1 1 0 0 0 0
B 0 0 0 0 0 0 1 1 1 1
C 0 0 1 0 0 0 0 1 0 0
D 0 1 0 1 1 0 1 0 0 1
E 1 0 1 1 0 1 0 0 1 0
F 0 1 0 0 0 0 0 0 1 0
G 1 0 1 0 0 0 0 0 0 0
H 0 1 0 1 0 0 1 0 0 1
I 1 0 1 0 0 0 0 1 1 0
I would like to create a sequence from two numbers, such that the occurrence of one of the numbers decreases (from n_1 to 1) while for the other number the occurrences are fixed at n_2.
I've been looking around for and tried using seq and rep to do it but I can't seem to figure it out.
Here is an example for c(0,1) and n_1=5, n_2=3:
0,0,0,0,0,1,1,1,0,0,0,0,1,1,1,0,0,0,1,1,1,0,0,1,1,1,0,1,1,1
And here for c(0,1) and n_1=2, n_2=1:
0,0,1,0,1
Maybe something like this?
rep(rep(c(0, 1), n_1), times = rbind(n_1:1, n_2))
## [1] 0 0 0 0 0 1 1 1 0 0 0 0 1 1 1 0 0 0 1 1 1 0 0 1 1 1 0 1 1 1
Here it is as a function (without any sanity checks):
myfun <- function(vec, n1, n2) rep(rep(vec, n1), times = rbind(n1:1, n2))
myfun(c(0, 1), 2, 1)
## [1] 0 0 1 0 1
inverse.rle
Another alternative is to use inverse.rle:
y <- list(lengths = rbind(n_1:1, n_2),
values = rep(c(0, 1), n_1))
inverse.rle(y)
## [1] 0 0 0 0 0 1 1 1 0 0 0 0 1 1 1 0 0 0 1 1 1 0 0 1 1 1 0 1 1 1
An alternative (albeit slower) method using a similar concept:
unlist(mapply(rep,c(0,1),times=rbind(n_1:1,n_2)))
###[1] 0 0 0 0 0 1 1 1 0 0 0 0 1 1 1 0 0 0 1 1 1 0 0 1 1 1 0 1 1 1
Here is another approach using upper-triangle of a matrix:
f_rep <- function(num1, n_1, num2, n_2){
m <- matrix(rep(c(num1, num2), times=c(n_1+1, n_2)), n_1+n_2+1, n_1+n_2+1, byrow = T)
t(m)[lower.tri(m,diag=FALSE)][1:sum((n_1:1)+n_2)]
}
f_rep(0, 5, 1, 3)
#[1] 0 0 0 0 0 1 1 1 0 0 0 0 1 1 1 0 0 0 1 1 1 0 0 1 1 1 0 1 1 1
f_rep(2, 4, 3, 3)
#[1] 2 2 2 2 3 3 3 2 2 2 3 3 3 2 2 3 3 3 2 3 3 3
myf = function(x, n){
rep(rep(x,n[1]), unlist(lapply(0:(n[1]-1), function(i) n - c(i,0))))
}
myf(c(0,1), c(5,3))
#[1] 0 0 0 0 0 1 1 1 0 0 0 0 1 1 1 0 0 0 1 1 1 0 0 1 1 1 0 1 1 1
I have a matrix that is for example like this:
rownames V1
a 1
c 3
b 2
d 4
y 2
q 4
i 1
j 1
r 3
I want to make a Symmetric binary matrix that it's dimnames of that is the same as rownames of above matrix. I want to fill these matrix by 1 & 0 in such a way that 1 indicated placing variables that has the same number in front of it and 0 for the opposite situation.This matrix would be like
dimnames
a c b d y q i j r
a 1 0 0 0 0 0 1 1 0
c 0 1 0 0 0 0 0 0 1
b 0 0 1 0 1 0 0 0 0
d 0 0 0 1 0 1 0 0 0
y 0 0 1 0 1 0 0 0 0
q 0 0 0 1 0 1 0 0 0
i 1 0 0 0 0 0 1 1 0
j 1 0 0 0 0 0 1 1 0
r 0 1 0 0 0 0 0 0 1
Anybody know how can I do that?
Use dist:
DF <- read.table(text = "rownames V1
a 1
c 3
b 2
d 4
y 2
q 4
i 1
j 1
r 3", header = TRUE)
res <- as.matrix(dist(DF$V1)) == 0L
#alternatively:
#res <- !as.matrix(dist(DF$V1))
#diag(res) <- 0L #for the first version of the question, i.e. a zero diagonal
res <- +(res) #for the second version, i.e. to coerce to an integer matrix
dimnames(res) <- list(DF$rownames, DF$rownames)
# 1 2 3 4 5 6 7 8 9
#1 1 0 0 0 0 0 1 1 0
#2 0 1 0 0 0 0 0 0 1
#3 0 0 1 0 1 0 0 0 0
#4 0 0 0 1 0 1 0 0 0
#5 0 0 1 0 1 0 0 0 0
#6 0 0 0 1 0 1 0 0 0
#7 1 0 0 0 0 0 1 1 0
#8 1 0 0 0 0 0 1 1 0
#9 0 1 0 0 0 0 0 0 1
You can do this using table and crossprod.
tcrossprod(table(DF))
# rownames
# rownames a b c d i j q r y
# a 1 0 0 0 1 1 0 0 0
# b 0 1 0 0 0 0 0 0 1
# c 0 0 1 0 0 0 0 1 0
# d 0 0 0 1 0 0 1 0 0
# i 1 0 0 0 1 1 0 0 0
# j 1 0 0 0 1 1 0 0 0
# q 0 0 0 1 0 0 1 0 0
# r 0 0 1 0 0 0 0 1 0
# y 0 1 0 0 0 0 0 0 1
If you want the row and column order as they are found in the data, rather than alphanumerically, you can subset
tcrossprod(table(DF))[DF$rownames, DF$rownames]
or use factor
tcrossprod(table(factor(DF$rownames, levels=unique(DF$rownames)), DF$V1))
If your data is large or sparse, you can use the sparse matrix algebra in xtabs, with similar ways to change the order of the resulting table as before.
Matrix::tcrossprod(xtabs(data=DF, ~ rownames + V1, sparse=TRUE))
I have a data set that looks like this:
Person Team
114 1
115 1
116 1
117 1
121 1
122 1
123 1
214 2
215 2
216 2
217 2
221 2
222 2
223 2
"Team" ranges from 1 to 33, and teams vary in terms of size (i.e., there can be 5, 6, or 7 members, depending on the team). I need to create a data set into something that looks like this:
1 1 1 1 1 1 1 0 0 0 0 0 0 0
1 1 1 1 1 1 1 0 0 0 0 0 0 0
1 1 1 1 1 1 1 0 0 0 0 0 0 0
1 1 1 1 1 1 1 0 0 0 0 0 0 0
1 1 1 1 1 1 1 0 0 0 0 0 0 0
1 1 1 1 1 1 1 0 0 0 0 0 0 0
1 1 1 1 1 1 1 0 0 0 0 0 0 0
0 0 0 0 0 0 0 1 1 1 1 1 1 1
0 0 0 0 0 0 0 1 1 1 1 1 1 1
0 0 0 0 0 0 0 1 1 1 1 1 1 1
0 0 0 0 0 0 0 1 1 1 1 1 1 1
0 0 0 0 0 0 0 1 1 1 1 1 1 1
0 0 0 0 0 0 0 1 1 1 1 1 1 1
0 0 0 0 0 0 0 1 1 1 1 1 1 1
The sizes of the individual blocks are given by the number of people in a team. How can I do this in R?
You could use bdiag from the package Matrix. For example:
> bdiag(matrix(1,ncol=7,nrow=7),matrix(1,ncol=7,nrow=7))
Another idea, although, I guess this is less efficient/elegant than RStudent's:
DF = data.frame(Person = sample(100, 21), Team = rep(1:5, c(3,6,4,5,3)))
DF
lengths = tapply(DF$Person, DF$Team, length)
mat = matrix(0, sum(lengths), sum(lengths))
mat[do.call(rbind,
mapply(function(a, b) arrayInd(seq_len(a ^ 2), c(a, a)) + b,
lengths, cumsum(c(0, lengths[-length(lengths)])),
SIMPLIFY = F))] = 1
mat
Say you have a matrix M1 as such:
A B C D E F G H I J
353 1 0 1 0 0 1 0 0 1 1
288 1 0 1 0 0 1 1 0 1 1
275 1 0 1 0 1 1 0 0 1 1
236 0 0 1 0 0 1 0 0 1 1
235 0 0 1 0 0 1 1 0 1 1
227 1 0 1 0 1 1 1 0 1 1
the rownames are the values (they are not random they have meaning and it is what I want as I will explain).
Say you have another matrix M2 as such:
A B C D E F G H I J AA
[1,] 0 0 0 0 0 0 0 0 0 0 0
[2,] 1 0 0 0 0 0 0 0 0 0 0
[3,] 0 1 0 0 0 0 0 0 0 0 1
[4,] 1 1 0 0 0 0 0 0 0 0 0
[5,] 0 0 1 0 0 0 0 0 0 0 1
[6,] 1 0 1 0 0 0 0 0 0 0 0
Note A to J is the same number of cols, except the 2 new cols, AA
Now, I want something like:
for (i in 1:nrow(M2)){
if(M2[i,"AA"]==1){
#-1 since I M1 doesnt have the AA column
vec = M2[i,1:(ncol(M2)-1)]
#BELOW is what I am not sure of what how to implement
#get the rowname from M1 that matches vec, and replace M2[i,"AA"] = that value
}
}
The result should be 0, since in this example there are no rows of M1 matching any rows of M2[,A:J]