vignettes/code/02-byte_f1.R

In jr-packages/efficientTutorial: Efficient R Programming

# dd = NULL
# for(i in seq(2, 4, length.out=12)) {
#   x = rnorm(10^i)
#   dd_tmp = rbenchmark::benchmark(my_mean(x), cmp_mean(x), mean(x),
#                                  columns = c("test", "elapsed", "relative"),
#                                  order = "relative", replications = 5000)
#   dd_tmp$i = i
#   dd = rbind(dd, dd_tmp)
# }
# dd$p = 10^dd$i
# dir.create("data", showWarnings = FALSE)
# mean_comparison = dd
# save(mean_comparison, file = "extdata/mean_comparison.RData")




source("code/initialise.R")
load(file = "extdata/mean_comparison.RData")

dd = mean_comparison
par(mar = c(3, 3, 2, 1), mgp = c(2, 0.4, 0), tck = -.01,
    cex.axis = 0.9, las = 1, xaxs = "i", yaxs = "i")
plot(0, type = "n", ylim = c(0, 200), xlim = c(10^2, 10^4),
     axes = FALSE, frame = FALSE,
     xlab = "Sample size", ylab = "Relative timings",
     log = "x")
abline(h = seq(0, 200, 50), lty = 1, col = "grey90")

m1 = dd[dd$test == "mean(x)", ]
lines(m1$p, m1$relative, lwd = 3, col = 3)
m1 = dd[dd$test == "cmp_mean(x)", ]
lines(m1$p, m1$relative, col = 2, lwd = 3)
m1 = dd[dd$test == "my_mean(x)", ]
lines(m1$p, m1$relative, col = 5, lwd = 3)

## Axis labels
axis(2, tick = FALSE,  col.axis = "grey50", cex.axis = 0.8)
axis(1, at = 10 ^ (2:4), labels = c(expression(10^2), expression(10^3),
                                expression(10^4)),
     tick = FALSE, col.axis = "grey50", cex.axis = 0.8)

## Title
title("Compiled vs non-compiled", adj = 1,
      cex.main = 0.9, font.main = 2, col.main = "black")

## Line labels
text(1000, 90, "Pure R", col = 5, lwd = 2, font = 2)
text(1000, 20, "Compiled R", col = 2, font = 2)
text(8000, 10, "mean", col = 3, font = 2)