Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Add code for #292 #293

Open
wants to merge 1 commit into
base: master
Choose a base branch
from
Open
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
1 change: 0 additions & 1 deletion 05-input-output.Rmd
Original file line number Diff line number Diff line change
Expand Up @@ -33,7 +33,6 @@ We also use the **WDI** package to illustrate accessing online data sets
library("WDI")
```


## Top 5 tips for efficient data I/O

1. If possible, keep the names of local files downloaded from the internet or copied onto your computer unchanged. This will help you trace the provenance of the data in the future.
Expand Down
113 changes: 55 additions & 58 deletions code/05-io_f1.R
Original file line number Diff line number Diff line change
@@ -1,39 +1,43 @@
source("code/initialise.R")
library(ggplot2)
library(data.table)
library(readr)
library("ggplot2")
library("data.table")
library("readr")
library("vroom")
library("arrow")
library("microbenchmark")

# # Start corresponds to 0.1 MB
# rows = 10^(seq(3.445,6, length.out = 50))
# cols = 2 * 10^(0:2)
# res = NULL
# for(i in seq_along(rows)) {
# for(k in 1:80) {
# for(j in seq_along(cols)) {
# no_of_rows = floor(rows[i]/(10^(j-1)))
# m = matrix(runif(no_of_rows * cols[j]), nrow = no_of_rows, ncol = cols[j])
# fname = tempfile()
# write.csv(m, file = fname, row.names = FALSE)
# mb = microbenchmark(times = 10,
# base_default = read.csv(fname),
# readr_default = read_csv(fname),
# fread_default = fread(fname)
# )
#
#
# tab = tapply(mb$time/1000, mb$expr, mean)
# res_tmp = data.frame(exp = names(tab),
# time = as.vector(tab), rows = no_of_rows,
# cols = cols[j])
# unlink(fname)
# res = rbind(res, res_tmp)
# save(res, file="04_tmp.RData")
# }
# }
# message(i)
# }
#save(res, file="extdata/04-f3.RData")
# Start corresponds to 0.1 MB
rows = 10^(seq(3.445, 4, length.out = 5))
cols = 2 * 5^(0:2)
res = NULL
for(i in seq_along(rows)) {
for(k in 1:3) {
for(j in seq_along(cols)) {
no_of_rows = floor(rows[i]/(10^(j-1)))
m = matrix(runif(no_of_rows * cols[j]), nrow = no_of_rows, ncol = cols[j])
fname = tempfile()
write.csv(m, file = fname, row.names = FALSE)
mb = microbenchmark(times = 10,
base_default = read.csv(fname),
readr_default = read_csv(fname),
fread_default = fread(fname),
vroom_default = vroom(fname, delim = ","),
arrow_default = arrow::read_csv_arrow(fname)
)


tab = tapply(mb$time/1000, mb$expr, mean)
res_tmp = data.frame(exp = names(tab),
time = as.vector(tab), rows = no_of_rows,
cols = cols[j])
unlink(fname)
res = rbind(res, res_tmp)
save(res, file="04_tmp.RData")
}
}
message(i)
}
save(res, file="extdata/04-f3.RData")
load("extdata/05-f1.RData")
res = aggregate(time ~ cols+rows+ exp, mean, data=res)
res$MB = res$cols*res$rows*18/1000000 ## Approximate
Expand All @@ -43,32 +47,32 @@ for(i in res$cells){
sel = res$cells == i
res$Time[sel] = res$time[sel] / min(res$time[sel])
}
res$type = factor(res$exp, labels=c("base", "data.table","readr"))
res$type = factor(res$exp, labels=c("base", "data.table","readr", "vroom"))

library("ggplot2")
res$facet_cols = paste("No of columns:", res$cols)
res = res[res$MB >= 0.1,]
g = ggplot(res, aes(MB, Time)) +
geom_line(aes(colour = type, linetype=type), size=1) +
g = ggplot(res, aes(MB, Time)) +
geom_line(aes(colour = type, linetype=type), size=1) +
facet_grid(~ facet_cols) +
scale_x_continuous(limits=c(min(res$MB),36), expand = c(0, 0), trans="log10") +
theme(panel.grid.major.y = element_line(colour = "gray90"),
panel.grid.minor = element_line(colour = NA),
panel.grid.major.x = element_line(colour = NA),
plot.title = element_text(size = 12,
face = "bold", hjust = 1, vjust = 0),
panel.background = element_rect(fill = NA),
legend.background = element_rect(fill = NA),
legend.position = c(0.95, 0.92),
scale_x_continuous(limits=c(min(res$MB),36), expand = c(0, 0), trans="log10") +
theme(panel.grid.major.y = element_line(colour = "gray90"),
panel.grid.minor = element_line(colour = NA),
panel.grid.major.x = element_line(colour = NA),
plot.title = element_text(size = 12,
face = "bold", hjust = 1, vjust = 0),
panel.background = element_rect(fill = NA),
legend.background = element_rect(fill = NA),
legend.position = c(0.95, 0.92),
axis.ticks.x = element_line(linetype = "blank"),
axis.ticks.y = element_line(linetype = "blank"),
legend.text = element_text(size = 11),
legend.text = element_text(size = 11),
legend.key = element_rect(fill = NA)) +
ylab("Relative time") + xlab("File size (MB)") +
scale_colour_manual(values=c(get_col(2), get_col(3), get_col(4))) +
scale_y_continuous(limits=c(0,15), expand = c(0, 0))
g1 = g + theme(strip.background = element_rect(fill = "white"),
strip.text = element_text( hjust = 0.95, face="bold")) +
ylab("Relative time") + xlab("File size (MB)") +
scale_colour_manual(values=c(get_col(2), get_col(3), get_col(4))) +
scale_y_continuous(limits=c(0,15), expand = c(0, 0))
g1 = g + theme(strip.background = element_rect(fill = "white"),
strip.text = element_text( hjust = 0.95, face="bold")) +
guides(colour = FALSE, linetype=FALSE)


Expand All @@ -82,10 +86,3 @@ labels = tibble::frame_data(
g2 = g1 + geom_text(data=labels, aes(color=type, label=type))
print(g2)