Construct a Global MVPA Result
Usage
global_mvpa_result(
performance_table,
result,
importance_map,
importance_vector,
activation_patterns,
raw_weights,
fold_fits,
model_spec
)Arguments
- performance_table
Tibble of cross-validated performance metrics.
- result
The merged classification/regression result object.
- importance_map
Spatial object (NeuroVol/NeuroSurface) of per-feature importance.
- importance_vector
Numeric vector of per-feature importance.
- activation_patterns
The P x D activation pattern matrix A.
- raw_weights
The averaged P x D weight matrix W.
- fold_fits
Optional list of per-fold model_fit objects.
- model_spec
The input model specification.
Examples
if (FALSE) { # \dontrun{
# Typically created by run_global(), not directly
result <- global_mvpa_result(
performance_table = tibble::tibble(Accuracy = 0.8),
result = NULL,
importance_map = NULL,
importance_vector = c(0.1, 0.2),
activation_patterns = NULL,
raw_weights = NULL,
fold_fits = NULL,
model_spec = list()
)
} # }