R tidyverse useful tricks

tidyverse.R

library(tidyverse)

data <- tibble::tibble(
  x = c(1, 2, NA, 2, NA), 
  y = c("a", NA, "b", NA, "b"),
  z = c("x", NA, "y", NA, NA)
)

# Get counts of NA values for each column in a data frame
data %>%
  purrr::map_df(~sum(is.na(.))) %>%
  tidyr::pivot_longer(cols = everything()) %>%
  dplyr::rename(column = name,
         na_count = value) %>%
  dplyr::arrange(-na_count)

# Remove rows with NA value present in any of the columns
data %>%
  tidyr::drop_na()
  
# Replace NA values in all columns
data %>%
  dplyr::mutate_all(~replace(., is.na(.), "N/A")) 

# Replace NA values in all columns of the same data type, ex. is.character
# either mutate_if or mutate + across
data %>%
  dplyr::mutate_if(is.character, ~tidyr::replace_na(., "N/A")) 
  
data %>%
  dplyr::mutate(
    dplyr::across(where(is.character), \(x) tidyr::replace_na(x, "N/A"))
  )
  
# Transform named list to tibble
my_list <- list(x = 5, y = 4, z = 12)

df <- my_list %>%
  tibble::enframe() %>%
  tidyr::unnest(cols = value)

# Create an empty data frame
df <- data.frame(
  doubles = double(),
  ints = integer(),
  factors = factor(),
  logicals = logical(),
  characters = character(),
  stringsAsFactors=FALSE
  )

# Simple Stack Overflow Questions

# Drop column by name
data %>%
  dplyr::select(-z, -y)