Runs guess_data_type() on every column in data, times each call, and
scores the predicted $route against the gold label at every ontology level
under hierarchical partial credit (a level counts only if all its ancestors
match). gold_style / gold_unit are scored only where defined.
Usage
fx_benchmark(
data = fx_benchmark_data(),
router = "auto",
weights = .fx_level_weights(),
max_n = 300L,
progress = FALSE
)Arguments
- data
A benchmark data frame from
fx_benchmark_data().- router
Router strategy passed to
guess_data_type()("auto","learned","signature").- weights
Named partial-credit weights over the ontology levels.
- max_n
Cap on the reconstructed column length (speed guard).
- progress
If
TRUE, print a progress dot every 50 cases.