Adding {data.table} tutorial

This commit is contained in:
Jose 2022-12-29 16:44:58 -03:00
parent 372a0896a5
commit d46e60c05c
4 changed files with 271 additions and 0 deletions

View File

@ -40,3 +40,9 @@ data using the R programming environment
* [[./script/roadoi_package.R]['roadoi']]
* [[./script/scraping.R]['crul' package - web scraping]]
** Data wrangling with {data.table}
* [[./doc/datatable_intro.org][Introduction to {data.table}]]
* [[./doc/datatable_querying.org][Querying in {data.table}: filter in ~i~ ]]
* [[./doc/datatable_querying_j.org][Querying in {data.table}: select in ~j~ ]]

90
doc/datatable_intro.org Normal file
View File

@ -0,0 +1,90 @@
#+TITLE: Datatable package
#+DATE: 2021-10-26
#+OPTIONS: creator:nil timestamp:nil todo:nil num:nil
#+PROPERTY: header-args:R :results output :session *Rc* :cmdline :tangle yes
#+PROPERTY: header-args:R+ :exports both
#+SETUPFILE: https://fniessen.github.io/org-html-themes/org/theme-readtheorg.setup
#+HTML_HEAD: <style> #content{max-width:1800px;}</style>
#+HTML_HEAD: <style> p{max-width:800px;}</style>
#+HTML_HEAD: <style> li{max-width:800px;}</style>
#+HTML_HEAD: <style> pre.src{background: #f5f5f5;color:black;}</style>
#+HTML_HEAD: <style> .example{background: #fefefa;color:black;}</style>
#+begin_export html
<div id="subtitle" style="float: center; text-align: center;">
<p>
Org document with
<a href="http://www.r-project.org/">R</a> code
</p>
<p>
<a href="http://www.r-project.org/">
<img src="http://www.r-project.org/Rlogo.jpg"/>
</a>
</p>
</div>
#+end_export
#+begin_src R :exports code
rm(list = ls())
#+end_src
** Introduction
#+begin_src R
library(data.table)
#+end_src
Reading data into R using ~{data.table}~
#+begin_src R :exports code
url = "https://raw.githubusercontent.com/Rdatatable/data.table/master/vignettes/flights14.csv"
flights <- fread(url)
#+end_src
#+RESULTS:
#+begin_src R
names(flights)
#+end_src
#+begin_src R
flights
#+end_src
#+begin_src R
nrow(flights)
#+end_src
#+RESULTS:
: [1] 253316
Creating a data.table:
#+begin_src R
dt <- data.table(col1 = c(1:6),
col2 = LETTERS[1:6],
col3 = (1))
#+end_src
#+begin_src R
class(dt)
#+end_src
#+begin_src R
class(dt$col2)
#+end_src
** Some references to learn more about {data.table}
* https://cran.r-project.org/web/packages/data.table/vignettes/datatable-intro.html
* See vignettes in R:
: > vignette(package="data.table")
* https://atrebas.github.io/post/2019-03-03-datatable-dplyr/
* https://rdatatable.gitlab.io/data.table/
* https://s3.amazonaws.com/assets.datacamp.com/blog_assets/datatable_Cheat_Sheet_R.pdf
* https://stackoverflow.com/questions/21435339/data-table-vs-dplyr-can-one-do-something-well-the-other-cant-or-does-poorly
* ~dtplyr~ package:
* https://github.com/tidyverse/dtplyr

View File

@ -0,0 +1,67 @@
#+TITLE: Datatable package
#+DATE: 2021-10-26
#+OPTIONS: creator:nil timestamp:nil todo:nil num:nil
#+PROPERTY: header-args:R :results output :session *Rc* :cmdline :tangle yes
#+PROPERTY: header-args:R+ :exports both
#+SETUPFILE: https://fniessen.github.io/org-html-themes/org/theme-readtheorg.setup
#+HTML_HEAD: <style> #content{max-width:1800px;}</style>
#+HTML_HEAD: <style> p{max-width:800px;}</style>
#+HTML_HEAD: <style> li{max-width:800px;}</style>
#+HTML_HEAD: <style> pre.src{background: #f5f5f5;color:black;}</style>
#+HTML_HEAD: <style> .example{background: #fefefa;color:black;}</style>
#+begin_export html
<div id="subtitle" style="float: center; text-align: center;">
<p>
Org document with
<a href="http://www.r-project.org/">R</a> code
</p>
<p>
<a href="http://www.r-project.org/">
<img src="http://www.r-project.org/Rlogo.jpg"/>
</a>
</p>
</div>
#+end_export
#+begin_src R :exports code
rm(list = ls())
#+end_src
** Querying data.table
Sintaxis:
* ~DT[i, j, by]~
Subseting by "i" and "j"
** Subset rows in "i"
#+begin_src R
dt1 <- flights[origin == "JFK" & month == 6L]
dt1
#+end_src
Get the first two rows
#+begin_src R
flights[1:2]
#+end_src
#+RESULTS:
: year month day dep_delay arr_delay carrier origin dest air_time distance
: 1: 2014 1 1 14 13 AA JFK LAX 359 2475
: 2: 2014 1 1 -3 13 AA JFK LAX 363 2475
: hour
: 1: 9
: 2: 11
Ordering results of querying
#+begin_src R
dt_filt <- flights[order(origin, -dest)]
dt_filt
#+end_src

View File

@ -0,0 +1,108 @@
#+TITLE: Datatable package
#+DATE: 2021-10-26
#+OPTIONS: creator:nil timestamp:nil todo:nil num:nil
#+PROPERTY: header-args:R :results output :session *Rc* :cmdline :tangle yes
#+PROPERTY: header-args:R+ :exports both
#+SETUPFILE: https://fniessen.github.io/org-html-themes/org/theme-readtheorg.setup
#+HTML_HEAD: <style> #content{max-width:1800px;}</style>
#+HTML_HEAD: <style> p{max-width:800px;}</style>
#+HTML_HEAD: <style> li{max-width:800px;}</style>
#+HTML_HEAD: <style> pre.src{background: #f5f5f5;color:black;}</style>
#+HTML_HEAD: <style> .example{background: #fefefa;color:black;}</style>
#+begin_export html
<div id="subtitle" style="float: center; text-align: center;">
<p>
Org document with
<a href="http://www.r-project.org/">R</a> code
</p>
<p>
<a href="http://www.r-project.org/">
<img src="http://www.r-project.org/Rlogo.jpg"/>
</a>
</p>
</div>
#+end_export
#+begin_src R :exports code
rm(list = ls())
#+end_src
** Select columns in "j"
This creates a vector
#+begin_src R
dt_c <- flights[, arr_delay]
head(dt_c)
#+end_src
To create a data.table, use the function "~list~" or dot "~.~"
#+begin_src R
dt_c1 <- flights[, list(arr_delay)]
head(dt_c1)
#+end_src
The selection is wrapped in "~list~", thus, the columns can be
renamed
#+begin_src R
dt_c2 <- flights[, .(delay2arriv = arr_delay, dist = distance)]
head(dt_c2)
#+end_src
#+begin_src R
dt_c2a <- flights[, c("arr_delay", "distance")]
head(dt_c2a)
#+end_src
#+begin_src R
dt_c2b <- flights[, year:day]
head(dt_c2b)
#+end_src
Building new colummns
#+begin_src R
head(dt_c2b[, `:=`(time = year -1, # name = value,
pumpkin = month,
tomatoe = day +1)])
#+end_src
Renaming colummns
#+begin_src R
setnames(dt_c2b,
c("time", "pumpkin", "tomatoe"),# old names
c("time_minus1", "pumpkin_month", "day_plus1")) # new names
names(dt_c2b)
#+end_src
#+begin_src R
setnames(dt_c2b, toupper) # new names
names(dt_c2b)
#+end_src
Renaming colummns
#+begin_src R
setnames(dt_c2b, tolower) # new names
names(dt_c2b)
#+end_src
#+begin_src R
dt_c2c <- flights[, -(year:day)]
head(dt_c2c)
#+end_src
Select columns by patterns, e.g., selecting 5 lines form columns having "del"
characters in the name:
#+begin_src R
flights[1:5, names(flights) %like% "del", with = FALSE]
#+end_src