grayskripko

dbtbl |> filter(row_number() < .., .by=..)

grayskripko

expand_grid(
  pick_n = 1:3) |> 
  rowwise() |> 
  mutate(mtr = unlist(list(measure(
    sam_vos, n_weeks=4, fn = \(x) sort(x)[pick_n])))) |> 
  # OR unnamed
  mutate(mtr = unlist(list(apply_to_wur(lst(across(ev()))))) |>
  ungroup() |> 
  pprint()
# more: https://dplyr.tidyverse.org/articles/rowwise.html#repeated-function-calls

grayskripko

# you can use it in Power BI to build a table of insights that will help you not to miss important influencers
suppressWarnings(suppressPackageStartupMessages({
  library(tidyverse)
  library(forcats)
  library(ggplot2)
  library(Hmisc)}))

custom_lump <- function(fct, min_size=nrow(tbl) / n_bins) {
  kept_levels <- fct %>%
    table() %>%

grayskripko

tbl %>% mutate(res=pmap(., ~with(list(...), {any_col is available here!

grayskripko

?dplyr::context
# gives us
cur_column() - current column name!
cur_<press tab>
n()

disp <- 10
mtcars %>% mutate(disp = .data$disp * .env$disp)
mtcars %>% mutate(disp = disp * disp)
?.data

grayskripko

group_in_func <- function(x, ...) {
  x %>% 
    group_by(!!!(...)) %>% 
    summarise(sum(mpg))
}

list(quos(any_name=cyl == 6)) %>% 
  map(
    group_in_func, 
    x=head(select(as_tibble(mtcars), 1:2), 5) %>% print())

grayskripko

pivot_longer(
  cols=-c(date, match_id),
  names_to=c("side", ".value"),
  names_sep='(?<=^[ha])_')
      
pivot_wider(
  id_cols=c(date, h_team, a_team), 
  names_from=side, 
  names_glue = "{side}_{.value}",
  values_from=-c(date, h_team, a_team, team, side))

grayskripko

# let's say we have a set of features and want to create a bunch of their interations
# Which should we use? I tried to answer this question thinking that 
# correlation is a good measure of novelty and usefullness of the information

explore$interactions <- function(df) {
  assert(ncol(df) == 2)
  df %>% 
    # { print(explore$cor_tibble(.)); . } %>% 
    mutate(sum=.[[1]] + .[[2]]) %>% 
    mutate(dif=.[[1]] - .[[2]]) %>% 

grayskripko

tbl %>%
  rowwise() %>%
  group_map(~{...})
  
  
tbl %>%
  mutate(i=1:n()) %>% 
  nest(data=-i) %>% 
  pluck('data') %>% 
  map(~{...})

grayskripko

def inve(x):
    print(type(x), x)


s = pd.Series([1,2,3])
s.pipe(inve)  # series
s.apply(inve)  # int 1
s.transform(inve)  # int 1
s.map(inve)  # int 1