for(index in values){
code to be run
}Control Flow: Loops
We want to look at how to control the execution of our code. The three main things we are looking at here are
if/then/elselogic and syntax- looping to repeatedly execute code
- vectorized functions for improved efficiency
This section looks at how to do loops (repeated execution of code) in R.
Looping in R
There are a number of ways to do looping in R
for()while()repeat()
The idea of each is to run some code repeatedly; often changing something with each execution of the code.
For Loops
The syntax for a for loop (most commonly used loop in R) is
where
- index defines a ‘counter’ or variable that varies
- ‘values’ define which values the index takes on
For example, our index below is i and the values it can take on are the integers from 1 to 10 (1:10)
for (i in 1:10){
print(i)
}[1] 1
[1] 2
[1] 3
[1] 4
[1] 5
[1] 6
[1] 7
[1] 8
[1] 9
[1] 10The values don’t need to take on numbers and the object you use for the index can be changed:
for (index in c("cat","hat","worm")){
print(index)
}[1] "cat"
[1] "hat"
[1] "worm"Of course, the idea is to use the changing values in some meaningful way. Here is a quick example of printing out a particular string based on inputs.
Create two vectors of length 5.
words<-c("first", "second", "third", "fourth", "fifth")
data <- runif(5)- Loop through the elements of these and print out the phrase
“The (#ed) data point is (# from data vector).”
- To put character strings together with other
Robjects (which will be coerced to strings) we can use thepaste()function. Checking the help we see:
paste (..., sep = " ", collapse = NULL, recycle0 = FALSE)
where ... ‘is one or more R objects, to be converted to character vectors.’ and the sep = argument determines the value by which to separate these objects.
paste("The ", words[2], " data point is ", data[2], ".", sep = "&")[1] "The &second& data point is &0.258889200165868&."paste("The ", words[1], " data point is ", data[1], ".", sep = "")[1] "The first data point is 0.260954210069031."Note: sep = "" is equivalent to using the paste0() function.
Ok, let’s put this into a loop!
for (i in 1:5){
print(paste0("The ", words[i], " data point is ", data[i], "."))
}[1] "The first data point is 0.260954210069031."
[1] "The second data point is 0.258889200165868."
[1] "The third data point is 0.541771301766858."
[1] "The fourth data point is 0.945435663685203."
[1] "The fifth data point is 0.900441576261073."- As
iiterates from 1 to 5, we pull out the corresponding elements ofwordsanddatato make our sentence!
A more useful example would be finding summary statistics about different numeric columns of a data frame (recall this is a 2D structure we often use to store datasets).
Consider a dataset on batting of Major League Baseball (MLB) players.
- You may need to run
install.packages("Lahman")once on your machine before you can run this code
- You may need to run
library(Lahman)
my_batting <- Batting[, c("playerID", "teamID", "G", "AB", "R", "H", "X2B", "X3B", "HR")]
head(my_batting) playerID teamID G AB R H X2B X3B HR
1 abercda01 TRO 1 4 0 0 0 0 0
2 addybo01 RC1 25 118 30 32 6 0 0
3 allisar01 CL1 29 137 28 40 4 5 0
4 allisdo01 WS3 27 133 28 44 10 2 2
5 ansonca01 RC1 25 120 29 39 11 3 0
6 armstbo01 FW1 12 49 9 11 2 1 0- Let’s say we want to find the
summary()for each numeric column of this data set.
summary(my_batting[ , "G"]) Min. 1st Qu. Median Mean 3rd Qu. Max.
1.00 12.00 34.00 50.74 79.00 165.00 summary(my_batting[ , "AB"]) Min. 1st Qu. Median Mean 3rd Qu. Max.
0.0 4.0 46.0 139.2 224.0 716.0 That’s fine but we want to do it for all the numeric columns. Let’s use a for loop!
dim(my_batting)[1] 108789 9We could do a loop that takes on values of 3:9 (or programmatically 3:dim(my_batting)[2]).
for (i in 3:dim(my_batting)[2]){
print(summary(my_batting[ , i]))
} Min. 1st Qu. Median Mean 3rd Qu. Max.
1.00 12.00 34.00 50.74 79.00 165.00
Min. 1st Qu. Median Mean 3rd Qu. Max.
0.0 4.0 46.0 139.2 224.0 716.0
Min. 1st Qu. Median Mean 3rd Qu. Max.
0.00 0.00 4.00 18.48 27.00 198.00
Min. 1st Qu. Median Mean 3rd Qu. Max.
0.00 0.00 8.00 36.39 56.00 262.00
Min. 1st Qu. Median Mean 3rd Qu. Max.
0.000 0.000 1.000 6.202 9.000 67.000
Min. 1st Qu. Median Mean 3rd Qu. Max.
0.000 0.000 0.000 1.247 1.000 36.000
Min. 1st Qu. Median Mean 3rd Qu. Max.
0.00 0.00 0.00 2.85 2.00 73.00 Alternatively, the seq_along() function can be useful. This looks at the length of the object and creates a sequence from 1 to that length. Remember that a data frame is truly a list of equal length vectors (usually). The length of a list is number of elements. Here that is the number of columns!
length(my_batting)[1] 9seq_along(my_batting)[1] 1 2 3 4 5 6 7 8 9Now we can just remove the 1st and 2nd entries of that vector (as they are not numeric columns) and use that as our values to iterate across.
for (i in seq_along(my_batting)[-1:-2]){
print(summary(my_batting[ , i]))
} Min. 1st Qu. Median Mean 3rd Qu. Max.
1.00 12.00 34.00 50.74 79.00 165.00
Min. 1st Qu. Median Mean 3rd Qu. Max.
0.0 4.0 46.0 139.2 224.0 716.0
Min. 1st Qu. Median Mean 3rd Qu. Max.
0.00 0.00 4.00 18.48 27.00 198.00
Min. 1st Qu. Median Mean 3rd Qu. Max.
0.00 0.00 8.00 36.39 56.00 262.00
Min. 1st Qu. Median Mean 3rd Qu. Max.
0.000 0.000 1.000 6.202 9.000 67.000
Min. 1st Qu. Median Mean 3rd Qu. Max.
0.000 0.000 0.000 1.247 1.000 36.000
Min. 1st Qu. Median Mean 3rd Qu. Max.
0.00 0.00 0.00 2.85 2.00 73.00 We likely don’t enjoy this format. Although we’ll see much easier ways to deal with this, let’s initialize a data frame to store our results in. We can initialize the type of data to store in a particular column using character(), numeric(), logical(), etc.
summary_df <- data.frame(stat = character(),
min = numeric(),
Q1 = numeric(),
Median = numeric(),
Mean = numeric(),
Q3 = numeric(),
Max = numeric())
summary_df[1] stat min Q1 Median Mean Q3 Max
<0 rows> (or 0-length row.names)Ok, now let’s fill this in as we loop (note we use i-2 to start filling in at row 1 and we grab the statistic we are summarizing from the colnames of the my_batting data frame).
for (i in seq_along(my_batting)[-1:-2]){
summary_df[i-2, ] <- c(colnames(my_batting[i]),
summary(my_batting[ , i]))
}
summary_df stat min Q1 Median Mean Q3 Max
1 G 1 12 34 50.7404884685033 79 165
2 AB 0 4 46 139.24132035408 224 716
3 R 0 0 4 18.483495574001 27 198
4 H 0 0 8 36.3886054656261 56 262
5 X2B 0 0 1 6.20202410170146 9 67
6 X3B 0 0 0 1.24707461232294 1 36
7 HR 0 0 0 2.85015029093015 2 73While Loops
- These provide an alternative way to loop when we don’t necessarily know how many iterations to do before we start.
while(cond) {
expr
}- If
condisFALSEthen the loop never executes. - We won’t use these much.
Other Loop Things
- Sometimes we need to jump out of a loop.
breakkicks you out of the loop.
for (i in 1:5){
if (i == 3) break #can put code to execute on the same line
print(paste0("The ", words[i], " data point is ", data[i], "."))
}[1] "The first data point is 0.260954210069031."
[1] "The second data point is 0.258889200165868."- Sometimes we need to skip an iteration.
nextjumps to the next iteration of the loop.
for (i in 1:5){
if (i == 3) next
print(paste0("The ", words[i], " data point is ", data[i], "."))
}[1] "The first data point is 0.260954210069031."
[1] "The second data point is 0.258889200165868."
[1] "The fourth data point is 0.945435663685203."
[1] "The fifth data point is 0.900441576261073."Quick R Video
Please pop this video out and watch it in the full panopto player!
Recap!
- Loops provide a mechanism to run the same code repeatedly
for(index in values){
#code to evaluate
}index is the variable that changes during each iteration
values are the values the index takes on
Use the table of contents on the left or the arrows at the bottom of this page to navigate to the next learning material!