Automated Exploratory Analysis
(auto EDA) using the report()
function of the explore
package.
In this example we use synthetic data created by the
create_data_churn()
function of the explore package.
Then we generate the report.
#> 1 000 (1k) observations with 9 variables
#> 0 observations containing missings (NA)
#> 0 variables containing missings (NA)
#> 0 variables with no variance
#> # A tibble: 9 × 8
#> variable type na na_pct unique min mean max
#> <chr> <chr> <int> <dbl> <int> <dbl> <dbl> <dbl>
#> 1 price dbl 0 0 26 4 19.1 29
#> 2 type chr 0 0 3 NA NA NA
#> 3 usage dbl 0 0 160 0 54.7 150
#> 4 shared int 0 0 2 0 0.4 1
#> 5 device chr 0 0 3 NA NA NA
#> 6 newsletter int 0 0 2 0 0.5 1
#> 7 language chr 0 0 4 NA NA NA
#> 8 duration int 0 0 101 0 49.6 100
#> 9 churn dbl 0 0 2 0 0.35 1