Useful R code snippets that I often look for

rtips.R

# --------- RStudio -------------------
# knit from CLI
rmarkdown::render("test.Rmd", "html_document") 

# --------- data manipulation ---------
# wide-to-long = gather, long-to-wide = spread
wide %>%
  gather(key = column_name_of_column_names, 
         value = column_name_of_column_data,
         -allvariables, -whichwewant, -ineachrow)
long %>%
  spread(key = column_name_of_column_names, 
         value = column_name_of_column_data)

# --------- files and folders ---------
## find all files in folder 
files <- dir("folder", pattern = "*.xlsx")

## load xlsx files based on the list
d <- 
  files %>% 
  file.path("folder", .) %>% 
  map(read_excel) %>% 
  bind_rows(.id = "id")

## using nest for collapsed data
d <-
  participants %>%
  group_by(id) %>%
  do(nest(read_excel(.$filename)))

# --------- nested frames and models (purrr, broom) ---------
## nest (collapse and expand back)
dg <- d %>% group_by(id) %>% nest()
dg %>% unnest()

## calculate (no need for group_by)
models <- dg %>% 
  mutate(model1 = purrr::map(data, ~ lm(rt ~ distance, data = .))) %>% 
  mutate(model2 = purrr::map(data, ~ lm(rt ~ size, data = .))) %>% 
  gather(key = question, value = model, -data, -id)

## see results
models %>% unnest(model %>% purrr::map(broom::tidy))
models %>% unnest(model %>% purrr::map(broom::glance))

## without nesting, also: http://varianceexplained.org/r/broom-intro/
### general summary
d %>% 
  group_by(id) %>% 
  do(broom::glance(lm(rt ~ x, data = .)))
### individual predictors/coefs
d %>% 
  group_by(id) %>% 
  do(broom::tidy(lm(rt ~ x, data = .)))