-
Notifications
You must be signed in to change notification settings - Fork 5
/
900_utilities.Rmd
443 lines (250 loc) · 10.7 KB
/
900_utilities.Rmd
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
# (PART) REFERENCE {-}
<!--
This file by Martin Monkman
is licensed under a Creative Commons Attribution 4.0 International License
https://creativecommons.org/licenses/by/4.0/
-->
# Utilities {#utilities}
```{r echo = FALSE}
library(knitr)
opts_chunk$set(message = FALSE, warning = FALSE, cache = TRUE)
options(width = 100, dplyr.width = 100)
library(tidyverse)
library(datapasta)
library(glue)
library(here)
library(janitor)
```
One of the many great things about the R ecosystem is that data scientists around the world are tackling the same types of things that you and I might encounter, creating solutions to those problems, and then sharing those solutions as packages that add new functionality to the ecosystem.
This chapter introduces some of the packages that I have found useful.
## `{datapasta}` .
Reference page: https://milesmcbain.github.io/datapasta/
> datapasta is about reducing resistance associated with copying and pasting data to and from R. It is a response to the realisation that I often found myself using intermediate programs like Sublime to munge text into suitable formats. Addins and functions in datapasta support a wide variety of input and output situations, so it (probably) “just works”. Hopefully tools in this package will remove such intermediate steps and associated frustrations from our data slinging workflows.
We don't use {datapasta} in this course, but the chances are high that someday soon you'll need to copy-and-paste data.
## `{glue}` ·
Reference page: https://glue.tidyverse.org/
> Glue offers interpreted string literals that are small, fast, and dependency-free. Glue does this by embedding R expressions in curly braces which are then evaluated and inserted into the argument string.
* You can use a function that you have installed but not loaded through the `library()` function, by putting the name of the package followed by the function you want to use. In this case, we are using the `glue()` function from the {glue} package, so our code looks like `glue::glue()`. The variable from our data frame gets put in braces, and the string returned is all within quotation marks.
Here's a simple example:
```{r}
glue('BIDA', '302')
```
We can also paste together variables, by putting the name inside curly braces `{}`:
```{r}
course <- "BIDA"
course_number <- 405
glue({course}, {course_number})
```
We can specify a character to separate the elements we are gluing, with `.sep = "<string>"`. In this example it is a space:
```{r}
glue('_The Mandolorian_',
'is a tv series on the Disney+ streaming service',
'that features Baby Yoda.', .sep = " ")
```
Glue together separate year (as a variable), month, and day values, separated by a hyphen "-", to create a single date string.
```{r}
# assign year to variable
this_year <- 2020
# solution
glue({this_year},
'02',
'29',
.sep = "-")
```
### `glue_data()` {#util-glue-data}
We can glue data in a {dplyr} pipe with the `glue_data()` function.
First, let's filter the {gapminder} data so we have a smaller piece to work with:
```{r}
gm_3 <- gapminder |>
filter(year == 2007,
country %in% c("Canada", "Japan", "New Zealand"))
gm_3
```
Now we can create multiple strings, one for each of the rows in the `gm_3` tibble:
```{r}
gm_3 |>
glue_data("{country} has a life expectancy of {lifeExp}.")
```
We can also apply functions within the curly braces:
```{r}
gm_3 |>
glue_data("{country} has a life expectancy of {round(lifeExp, 1)}.")
```
Introducing the `starwars` character data table:
```{r}
head(starwars)
```
Create a new variable in the data frame by gluing the variables `species` and `gender`.
```{r}
# solution
starwars |>
mutate(species_gender = glue("{species} - {gender}")) |>
select(name, species, gender, species_gender)
```
## `{here}` ·
Reference page: https://here.r-lib.org/
> The goal of the here package is to enable easy file referencing in project-oriented workflows. In contrast to using setwd(), which is fragile and dependent on the way you organize your files, here uses the top-level directory of a project to easily build paths to files.
The `here()` package is used as part of the Workflow chapter.
## `{janitor}` ·
Reference page: https://sfirke.github.io/janitor/
> The janitor package is an R package that has simple functions for examining and cleaning dirty data. It was built with beginning and intermediate R users in mind and is optimised for user-friendliness.
> The main janitor functions:
>* perfectly format data frame column names;
>
>* isolate partially-duplicate records; and
>
>* provide quick tabulations (i.e., frequency tables and crosstabs).
| function | action |
| :-- | :-- |
| **cleaning** | |
| `clean_names()` | handles problematic variable names |
| `remove_empty()` | removes any columns that are entirely empty and entire rows that are entirely empty |
| `excel_numeric_to_date()` | converts Excel date serial numbers to class `Date` |
| **data summaries** | |
| `tabyl()` | takes a vector and returns a frequency table |
| `adorn_()` | a family of functions for additional summary values |
### `clean_names()`
This is going to demonstrate {janitor} `clean_names()` function.
The function
* changes the capitalization to lower case
* replaces spaces and periods with underscores ("_")
* changes unusual characters (e.g. percent) with text variations
* removes non-standard characters (e.g. dollar signs)
```{r}
# Create a data.frame with dirty names
test_df <- as.data.frame(matrix(ncol = 7))
names(test_df) <- c(
# snake case
"firstName",
# non-standard characters
"ábc@!*",
# percent and parentheses
"% successful (2009)",
# repeated name, with period separator
"REPEAT.VARIABLE",
"REPEAT.VARIABLE",
# two variables with blank for name
"",
"")
test_df
```
Now we apply the `clean_names()` function:
```{r}
clean_df <- test_df |>
janitor::clean_names()
clean_df
```
Compare the cleaned variable names with those in the original dataframe. What's changed?
### another `clean_names()` example
In this example, we read in the valid rows from the "deaths.xlsx" file (part of the [{readxl} package] https://readxl.tidyverse.org/).
```{r}
deaths <- read_excel("data/deaths.xlsx", skip = 4, n_max = 10)
deaths
```
If we wanted to filter the cases where the individual had children, we have to surround the `Has kids` variable name in backticks, since there is a space. We also have to remember to capitalize the "H" (although this is less of an issue if we are using RStudio, as the autofill is not sensitive to case).
```{r}
deaths |>
filter(`Has kids` == TRUE)
```
Using the `clean_names()` function from {janitor} gives us a solution: it makes the variable names consistently lower-case, and with the spaces replaced by an underscore ("_").
```{r}
deaths <- deaths |>
janitor::clean_names()
deaths
```
This then means that our filter statement does not need backticks around the variable name.
```{r}
deaths |>
filter(has_kids == TRUE)
```
### `remove_empty()`
The {janitor} package also has functions to clean up dataframes that have empty rows and columns.
First, we'll read in an Excel file called "dirty_data.xlsx".
```{r}
roster_raw <- readxl::read_excel("data/dirty_data.xlsx") # available at http://github.com/sfirke/janitor
glimpse(roster_raw)
roster_raw
```
As we can see, this table has messy names and a messy structure. In the code below, we'll use the `clean_names()` and `remove_empty()` functions from {janitor}.
```{r}
roster <- roster_raw |>
clean_names() |>
remove_empty(c("rows", "cols"))
roster
```
### Other useful {janitor} functions
We can also extend the pipe with other cleaning functions. Note that `excel_numeric_to_date()` is another {janitor} function, while `coalesce()` is from {dplyr} (and is similar to the SQL function of the same name).
```{r}
roster <- roster_raw |>
clean_names() |>
remove_empty(c("rows", "cols")) |>
mutate(hire_date = excel_numeric_to_date(hire_date),
cert = coalesce(certification_9, certification_10)) |> # from dplyr
select(-certification_9, -certification_10) # drop unwanted columns
roster
```
* `coalesce()` finds the first non-missing value at each position
```{r}
x <- c(1, 2, NA, 4)
y <- c(NA, NA, 93, 94)
coalesce(x, y)
```
What happens if you put `y` first in the `coalesce()` function?
```{r}
# solution
coalesce(y, x)
# the last value in the sequence is pulled from `y` instead of `x`
```
### Duplicate records with `get_dupes()`
We can find the cases where there are duplicates in the data records. In this case, we are looking for duplicate names:
```{r}
roster |> get_dupes(first_name, last_name)
```
Find the cases where the subject is duplicated.
```{r}
# solution
roster |> get_dupes(subject)
```
### Simple tables with `tabyl`
The {janitor} package also offers some simple tabulations with the `tabyl()` function.
One variable returns counts and percentages:
```{r}
roster |>
tabyl(full_time)
```
If we use two variables, it takes the first one as the rows, and the second as the columns:
```{r}
roster |>
tabyl(full_time, subject)
```
And for three variables, it makes separate tables based on the third one, then creates tables like the two-variable version:
```{r}
roster |>
tabyl(full_time, subject, employee_status)
```
There's also an option to drop levels with zero cases:
```{r}
roster |>
tabyl(full_time, subject, employee_status, show_missing_levels = FALSE)
```
### `adorn` tables with totals and percentages
You can also add totals and percentages with the `adorn_()` functions. This example creates a fancy version of what we had before.
* Note: I found the naming of the totals a bit counter-intuitive. `adorn_totals("row")` creates a new row with the total for that column, while `adorn_totals("col")` is a new column with the row total.
```{r}
roster |>
tabyl(employee_status, full_time) |>
adorn_totals("col") |>
adorn_percentages() |>
adorn_pct_formatting() |>
adorn_ns() |>
adorn_title("combined")
```
See the [{janitor} reference page](http://sfirke.github.io/janitor/) for additional details on the table functions.
## `{waldo}` ·
Have you ever needed to compare the contents of two Excel files? The original and the one that you maybe accidentally changed? The [{waldo}](https://waldo.r-lib.org/) has functions to help with that comparison.
## But wait, there's more!
Here's a twitter thread on packages that people nominated as deserving more attention:
* https://twitter.com/WeAreRLadies/status/1369984394836910084
And a list of some other useful packages:
* Garrett Grolemund, ["Quick list of useful R packages"](https://support.rstudio.com/hc/en-us/articles/201057987-Quick-list-of-useful-R-packages) (2021-03-18)
-30-