Build Mock ROI Data for Plugin Tests — mock_roi

Constructs a minimal ROI payload compatible with fit_roi. This is useful for unit-testing plugin methods without running run_searchlight() or run_regional().

Usage

mock_roi_data(
  train_data = NULL,
  test_data = NULL,
  indices = NULL,
  n_train = 20L,
  n_features = 10L,
  n_test = 0L,
  seed = NULL,
  train_roi = NULL,
  test_roi = NULL
)

Arguments

train_data: Optional matrix of training data (observations x features).
test_data: Optional matrix of test data (observations x features).
indices: Optional integer feature indices. Defaults to seq_len(ncol(train_data)).
n_train: Number of training observations to simulate when train_data is NULL.
n_features: Number of features to simulate when train_data is NULL.
n_test: Number of test observations to simulate when test_data is NULL and n_test > 0.
seed: Optional RNG seed used only when simulating data.
train_roi: Optional ROI object to include in the returned list.
test_roi: Optional ROI object to include in the returned list.

Value

A list with fields train_data, test_data, indices, train_roi, and test_roi.