uncomment this code chunks

[EmilHvitfeldt]

feature-engineering-az/categorical-cleaning.qmd

Lines 126 to 192 in 53646e2

	```{r}
	#| label: elevators-dv_manufacturer-stringdist
	#| eval: false
	#| echo: false
	library(tidyverse)
	self_stringdist <- function(x) {
	x <- unique(x)
	out <- stringdist::stringdistmatrix(x, x) |>
	reshape2::melt()
	out$Var1 <- x[out$Var1]
	out$Var2 <- x[out$Var2]
	out[out$value != 0, ]
	}
	elevators::elevators_raw |>
	filter(!is.na(DV_MANUFACTURER)) |>
	relocate(DV_MANUFACTURER) |>
	pull(DV_MANUFACTURER) |>
	self_stringdist() |>
	arrange(value)
	```
	```{r}
	#| label: elevators-dv_manufacturer-table
	#| eval: false
	#| echo: false
	library(tidyverse)
	elevators::elevators_raw |>
	select(DV_MANUFACTURER) |>
	drop_na() |>
	filter(str_detect(DV_MANUFACTURER, "^O")) |>
	filter(str_detect(DV_MANUFACTURER, "OLD", negate = TRUE)) |>
	pull(DV_MANUFACTURER) |>
	table()
	```
	```{r}
	#| label: wine_ratings
	#| eval: false
	#| echo: false
	wine_ratings <- readr::read_csv("https://raw.githubusercontent.com/rfordatascience/tidytuesday/master/data/2019/2019-05-28/winemag-data-130k-v2.csv")
	library(tidyverse)
	wine_ratings |>
	dplyr::count(region_1) |>
	View()
	library(tidyverse)
	self_stringdist <- function(x) {
	x <- unique(x)
	out <- stringdist::stringdistmatrix(x, x) |>
	reshape2::melt()
	out$Var1 <- x[out$Var1]
	out$Var2 <- x[out$Var2]
	out[out$value != 0, ]
	}
	wine_ratings$region_1 |>
	self_stringdist() |>
	as_tibble() |>
	arrange(value)
	```