R/apply_fits.R
apply_fits.RdThis function applies a given function to all
fit objects and returns the result in a tidy
tibble. Any function or purrr-style
lambda function can be used.
apply_fits(obj, .f, ..., .progress = FALSE, .options = furrr_options())a simpr_tibble with
repetition number, metaparameters, simulated
data, and fitted models, from
fit
A function or purrr-style
lambda function (see
as_mapper) used for
computing on the fit object
Additional arguments to .f.
A logical, for whether or not to print a progress bar for multiprocess, multisession, and multicore plans .
The future specific
options to use with the workers when using
futures. This must be the result from a call
to
furrr_options().
A tibble with columns .sim_id,
rep, Source (which contains the name of the
fit column), any metaparameters from
define, and additional columns
containing the results of .f applied to each fit
object.
set.seed(100)
logit_fit = specify(a = ~ sample(0:1, size = n, replace = TRUE),
b = ~ a + rlnorm(n)) %>%
define(n = c(40, 50)) %>%
generate(1) %>%
fit(logit = ~ glm(a ~ b, family = "binomial"))
logit_fit %>%
apply_fits(broom::augment)
#> # A tibble: 90 × 12
#> .sim_id n rep Source a b .fitted .resid .std.re…¹ .hat .sigma
#> <int> <dbl> <int> <chr> <int> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
#> 1 1 40 1 logit 1 1.14 -0.919 1.58 1.61 0.0304 1.12
#> 2 1 40 1 logit 1 1.88 -0.764 1.51 1.53 0.0259 1.12
#> 3 1 40 1 logit 0 0.653 -1.02 -0.784 -0.799 0.0370 1.15
#> 4 1 40 1 logit 1 4.26 -0.261 1.29 1.34 0.0713 1.13
#> 5 1 40 1 logit 1 1.18 -0.911 1.58 1.60 0.0300 1.12
#> 6 1 40 1 logit 0 1.64 -0.815 -0.856 -0.868 0.0265 1.14
#> 7 1 40 1 logit 0 6.10 0.127 -1.23 -1.36 0.174 1.13
#> 8 1 40 1 logit 0 1.46 -0.853 -0.843 -0.855 0.0276 1.14
#> 9 1 40 1 logit 1 1.33 -0.879 1.57 1.59 0.0285 1.12
#> 10 1 40 1 logit 0 1.82 -0.776 -0.870 -0.882 0.0259 1.14
#> # … with 80 more rows, 1 more variable: .cooksd <dbl>, and abbreviated variable
#> # name ¹.std.resid
## Arguments to the function can be passed in ...
logit_fit %>%
apply_fits(broom::augment, se_fit = TRUE)
#> # A tibble: 90 × 13
#> .sim_id n rep Source a b .fitted .se.fit .resid .std.r…¹ .hat
#> <int> <dbl> <int> <chr> <int> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
#> 1 1 40 1 logit 1 1.14 -0.919 0.386 1.58 1.61 0.0304
#> 2 1 40 1 logit 1 1.88 -0.764 0.345 1.51 1.53 0.0259
#> 3 1 40 1 logit 0 0.653 -1.02 0.436 -0.784 -0.799 0.0370
#> 4 1 40 1 logit 1 4.26 -0.261 0.539 1.29 1.34 0.0713
#> 5 1 40 1 logit 1 1.18 -0.911 0.383 1.58 1.60 0.0300
#> 6 1 40 1 logit 0 1.64 -0.815 0.353 -0.856 -0.868 0.0265
#> 7 1 40 1 logit 0 6.10 0.127 0.837 -1.23 -1.36 0.174
#> 8 1 40 1 logit 0 1.46 -0.853 0.363 -0.843 -0.855 0.0276
#> 9 1 40 1 logit 1 1.33 -0.879 0.371 1.57 1.59 0.0285
#> 10 1 40 1 logit 0 1.82 -0.776 0.347 -0.870 -0.882 0.0259
#> # … with 80 more rows, 2 more variables: .sigma <dbl>, .cooksd <dbl>, and
#> # abbreviated variable name ¹.std.resid
## Using a purrr-style lambda function
logit_fit %>%
apply_fits(~ summary(.)$cov.scaled)
#> # A tibble: 4 × 6
#> .sim_id n rep Source `(Intercept)` b
#> <int> <dbl> <int> <chr> <dbl> <dbl>
#> 1 1 40 1 logit 0.272 -0.0744
#> 2 1 40 1 logit -0.0744 0.0359
#> 3 2 50 1 logit 0.359 -0.167
#> 4 2 50 1 logit -0.167 0.102