Visualize importance scores

Usage

# S3 method for class 'importance_perm'
autoplot(
  object,
  top = Inf,
  metric = NULL,
  eval_time = NULL,
  type = "importance",
  std_errs = stats::qnorm(0.95),
  ...
)

Arguments

object: A tibble of results from importance_perm().
top: An integer for how many terms to show. To define importance when there are multiple metrics, the rankings of predictors are computed across metrics and the average rank is used. In the case of tied rankings, all the ties are included.
metric: A character vector or NULL for which metric to plot. By default, all metrics will be shown via facets. Possible options are the entries in .metric column of the object.
eval_time: For censored regression models, a vector of time points at which the survival probability is estimated.
type: A character value. The default is "importance" which shows the overall signal-to-noise ration (i.e., mean divided by standard error). Alternatively, "difference" shows the mean difference value with standard error bounds.
std_errs: The number of standard errors to plot (when type = "difference").
...: Not used.

Value

A ggplot2 object.

Examples

# Pre-computed results. See code at
system.file("make_imp_example.R", package = "important")
#> [1] "/home/runner/work/_temp/Library/important/make_imp_example.R"

# Load the results
load(system.file("imp_examples.RData", package = "important"))

# A classification model with two classes and highly correlated predictors.
# To preprocess them, PCA feature extraction is used.
#
# Let’s first view the importance in terms of the original predictor set
# using 50 permutations:

imp_orig
#> # A tibble: 390 × 6
#>    .metric     predictor     n    mean std_err importance
#>    <chr>       <chr>     <int>   <dbl>   <dbl>      <dbl>
#>  1 roc_auc     Genotype     50 0.108   0.00442      24.4 
#>  2 brier_class Genotype     50 0.0353  0.00151      23.3 
#>  3 accuracy    Genotype     50 0.0568  0.00314      18.1 
#>  4 roc_auc     Eotaxin_3    50 0.0568  0.00392      14.5 
#>  5 roc_auc     age          50 0.0622  0.00463      13.4 
#>  6 brier_class age          50 0.0190  0.00143      13.3 
#>  7 brier_class Eotaxin_3    50 0.0132  0.00130      10.2 
#>  8 accuracy    age          50 0.0183  0.00275       6.67
#>  9 roc_auc     male         50 0.0201  0.00352       5.72
#> 10 brier_class Resistin     50 0.00535 0.00115       4.65
#> # ℹ 380 more rows

autoplot(imp_orig, top = 10)


# Now assess the importance in terms of the PCA components

imp_derv
#> # A tibble: 36 × 6
#>    .metric     predictor         n   mean std_err importance
#>    <chr>       <chr>         <int>  <dbl>   <dbl>      <dbl>
#>  1 brier_class Genotype_E3E4    50 0.181  0.00332       54.5
#>  2 brier_class Genotype_E3E3    50 0.132  0.00293       45.0
#>  3 roc_auc     Genotype_E3E4    50 0.238  0.00632       37.6
#>  4 accuracy    Genotype_E3E4    50 0.140  0.00427       32.9
#>  5 brier_class Genotype_E2E3    50 0.0605 0.00196       30.9
#>  6 accuracy    Genotype_E3E3    50 0.0902 0.00364       24.7
#>  7 roc_auc     Genotype_E4E4    50 0.0841 0.00349       24.1
#>  8 roc_auc     Genotype_E3E3    50 0.106  0.00472       22.5
#>  9 brier_class Genotype_E4E4    50 0.0370 0.00166       22.3
#> 10 accuracy    Genotype_E2E3    50 0.0497 0.00260       19.1
#> # ℹ 26 more rows

autoplot(imp_derv)

autoplot(imp_derv, metric = "brier_class", type = "difference")