sample_draws: Sample draws from a tidy-format data frame of draws

In tidybayes: Tidy Data and 'Geoms' for Bayesian Models

Sample draws from a tidy-format data frame of draws

Description

Given a tidy-format data frame of draws with a column indexing each draw, subsample the data frame to a given size based on a column indexing draws, ensuring that rows in sub-groups of a grouped data frame are sampled from the same draws.

Usage

sample_draws(data, ndraws, draw = ".draw", seed = NULL)

Arguments

data	Data frame to sample from
ndraws	The number of draws to return, or NULL to return all draws.
draw	The name of the column indexing the draws; default ".draw".
seed	A seed to use when subsampling draws (i.e. when ndraws is not NULL).

Details

sample_draws() makes it easier to sub-sample a grouped, tidy-format data frame of draws. On a grouped data frame, the naive approach of using filter with the .draw column will give incorrect results as it will select a different sample within each group. sample_draws() ensures the same sample is selected within each group.

Author(s)

Matthew Kay

Examples

## Not run: 

library(ggplot2)
library(dplyr)
library(brms)
library(modelr)

theme_set(theme_light())

m_mpg = brm(mpg ~ hp * cyl, data = mtcars,
  # 1 chain / few iterations just so example runs quickly
  # do not use in practice
  chains = 1, iter = 500)

# draw 100 fit lines from the posterior and overplot them
mtcars %>%
  group_by(cyl) %>%
  data_grid(hp = seq_range(hp, n = 101)) %>%
  add_epred_draws(m_mpg) %>%
  # NOTE: only use sample_draws here when making spaghetti plots; for
  # plotting intervals it is always best to use all draws
  sample_draws(100) %>%
  ggplot(aes(x = hp, y = mpg, color = ordered(cyl))) +
  geom_line(aes(y = .epred, group = paste(cyl, .draw)), alpha = 0.25) +
  geom_point(data = mtcars)


## End(Not run)

tidybayes documentation built on Aug. 13, 2023, 1:06 a.m.