Construct a Region Importance Result

Usage

region_importance_result(
  importance,
  importance_map,
  p_values,
  p_value_map,
  stats_table,
  iteration_log,
  model_spec,
  importance_vector_raw = NULL
)

Arguments

importance: Numeric vector of per-feature importance scores.
importance_map: Spatial object of importance values.
p_values: Numeric vector of Wilcoxon p-values.
p_value_map: Spatial object of -log10(p) values.
stats_table: Tibble with per-feature statistics.
iteration_log: Tibble with per-iteration performance.
model_spec: The input model specification.
importance_vector_raw: Raw importance vector before spatial mapping (default NULL).

Value

An S3 object of class region_importance_result.

Examples

# Typically created by region_importance(), shown here with toy values
result <- region_importance_result(
  importance = c(0.1, 0.2),
  importance_map = NULL,
  p_values = c(0.05, 0.01),
  p_value_map = NULL,
  stats_table = tibble::tibble(feature_id = 1:2),
  iteration_log = tibble::tibble(iter = 1:10, performance = runif(10)),
  model_spec = list()
)
print(result)
#> 
#>   Region Importance Result
#> 
#>   Features: 2 ( 2 with valid importance)
#>   Iterations: 10 
#>   Importance range: [ 0.1 , 0.2 ]
#>   Top 2 features:
#>      2 : importance = 0.2 , p = 0.01 
#>      1 : importance = 0.1 , p = 0.05 
#>   Mean iteration performance: 0.4949 (SD: 0.2604 )
#>