Created
March 1, 2018 19:15
-
-
Save johnburnmurdoch/7374a41579b0fd56c2c8100e0aadb33f to your computer and use it in GitHub Desktop.
Vectorising for loops to save time in R
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
needs(tidyverse, magrittr, jsonlite, scales) | |
loopTimer <- function(rows){ | |
col1 <- runif (rows, 0, 2) | |
col2 <- rnorm (rows, 0, 2) | |
col3 <- rpois (rows, 3) | |
col4 <- rchisq (rows, 2) | |
df <- data.frame (col1, col2, col3, col4) | |
r <- system.time({ | |
for (i in 1:nrow(df)) { # for every row | |
if ((df[i, 'col1'] + df[i, 'col2'] + df[i, 'col3'] + df[i, 'col4']) > 4) { # check if > 4 | |
df[i, 5] <- "greater_than_4" # assign 5th column | |
} else { | |
df[i, 5] <- "lesser_than_4" # assign 5th column | |
} | |
} | |
}) | |
output <- character (nrow(df)) # initialize output vector | |
v <- system.time({ | |
for (i in 1:nrow(df)) { | |
if ((df[i, 'col1'] + df[i, 'col2'] + df[i, 'col3'] + df[i, 'col4']) > 4) { | |
output[i] <- "greater_than_4" | |
} else { | |
output[i] <- "lesser_than_4" | |
} | |
} | |
df$output}) | |
data.frame(type=names(r),Appending=as.numeric(r),Vectorised=as.numeric(v)) | |
} | |
rowNums <- c(1:10)^5 | |
timerList <- vector("list", length(rowNums)) | |
for(i in 1:length(rowNums)){ | |
n <- rowNums[i] | |
timerList[[i]] <- loopTimer(n) %>% mutate(nrow = n) | |
} | |
timerList %>% bind_rows %>% | |
filter(type == "elapsed") %>% | |
gather(method, seconds, 2:3) %>% | |
ggplot(aes(nrow, seconds, col=method, label = ifelse(nrow == max(nrow), paste0(method,' '), ''))) + | |
theme_minimal(base_family = "Avenir") + | |
theme( | |
legend.position = "none", | |
axis.line = element_line(color='black'), | |
panel.grid.minor = element_blank() | |
) + | |
geom_line() + | |
geom_text(hjust=1, vjust=0) + | |
scale_x_continuous(label = comma) + | |
scale_color_manual(values = c("Appending"="#0083eb", "Vectorised"="#ff2b4f")) + | |
labs( | |
title = "How to save time when processing data in for loops", | |
x = "Rows in data frame", | |
y = "Seconds elapsed" | |
) |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment