CODEWITHSUNDEEP
rloopsfor-looptime
stats_noob
stats_noob

Reputation: 5935

Is Sys.time() the time required for the individual computer to run commands?

I am using the R programming language. I am learning about different types of "times" in computing. There seems to be two types of general times : the time corresponding to the general mathematical complexity of the procedures you are running, and the time corresponding to how long it takes your individual computer to run the procedures.

I came across the Sys.time() command in R over here: https://www.r-bloggers.com/2017/05/5-ways-to-measure-running-time-of-r-code/

Does Sys.time() measure the time required for your individual computer to run the commands? For example, I wrote the code below using Sys.time() to measure the time it takes my personal computer to run the following procedure in R for different numbers of observations.

 library(Rtsne)
    library(cluster)
    library(ggplot2)
    library(dplyr)
    library(dbscan)
    library(plotly)


#generate data

var_1 <- rnorm(1000,1,4)
var_2<-rnorm(1000,10,5)
var_3 <- sample( LETTERS[1:4], 1000, replace=TRUE, prob=c(0.1, 0.2, 0.65, 0.05) )
response_variable <- sample( LETTERS[1:2], 100, replace=TRUE, prob=c(0.4, 0.6) )


#put them into a data frame called "f"
f <- data.frame(var_1, var_2, var_3, response_variable)

#declare var_3 and response_variable as factors
f$response_variable = as.factor(f$response_variable)
f$var_3 = as.factor(f$var_3)

#create id
f$ID <- seq_along(f[,1])


mseDF <- NULL

for (size in c(100,500,1000)) {
   
    start_time <- Sys.time()
   
   
    #gower distance
   
    gower_dist <- daisy(f[1:size, 1:4],
                        metric = "gower")
   
    lof <- lof(gower_dist, k=3)
   
    tsne_obj <- Rtsne(gower_dist,  is_distance = TRUE)
   
    tsne_data <- tsne_obj$Y %>%
        data.frame() %>%
        setNames(c("X", "Y")) %>%
        mutate(
            name = f[1:size,]$ID,
            lof=lof,
            var1=f[1:size,]$var_1,
            var2=f[1:size,]$var_2,
            var3=f[1:size,]$var_3
        )
   
    p1 <- ggplot(aes(x = X, y = Y, size=lof, key=name, var1=var1,
                     var2=var2, var3=var3), data = tsne_data) +
        geom_point(shape=1, col="red")+
        theme_minimal()
   
   
    ggplotly(p1, tooltip = c("lof", "name", "var1", "var2", "var3"))
    end_time <- Sys.time()
   
   (final = end_time - start_time)
   print (final)
mseDF <- rbind(mseDF, data.frame(size, final))

}

Is this correct?

Extra: visualize the run times:

plot(mseDF)
 lines(mseDF)

Thanks

Upvotes: 0

Views: 299

Answers (2)

Retired Data Munger
Retired Data Munger

Reputation: 1445

Another way is to use proc.time; e.g.

> start_time <- proc.time()
> # some computation
> for (i in 1:10000000) i ^ 3
> proc.time() - start_time
   user  system elapsed 
   0.70    0.02    0.72

Upvotes: 1

IRTFM
IRTFM

Reputation: 263479

Does Sys.time() measure the time required for your individual computer to run the commands?

No. Sys.time() returns the current time. Similarly Sys.Date() returns the current date. You could create an elapsed time function that has code in it by subtracting two calls to Sys.time. But your code implies that you already know how to do that, so perhaps English is not your first language and that question was a simple error in expression. So maybe "yes", since it appears you would get the elapsed time for a result to appear.

You might want to look at ?system.time

system.time({for(i in 1:10e6){x <- Sys.time()}})
#   user  system elapsed 
# 13.274   0.000  13.273

There are also several benchmark packages (as demonstrated in that blog citation above.). Example: pkg:microbenchmark

Upvotes: 1

Related Questions