Calibrate the Per-task Cost Model on the Current Machine

Runs a small, fixed benchmark sweep on the current architecture - tiny nboot values, all ssdtools::ssd_ci_methods(), and a couple of nrow values - times each ssdtools::ssd_hc() call, fits the per-ci_method cost model time = (base + slope * max(nboot, n0)) * nrow_factor(nrow), and returns a versioned ssdsims_cost_calibration object carrying the fitted coefficients and provenance. Producing an architecture-specific estimator is nothing more than this single call plus passing the result to ssd_estimate_cost().

Usage

ssd_calibrate_cost(
  nboot = c(20L, 50L, 100L, 200L),
  nrow = c(5L, 10L, 20L, 50L),
  data = NULL,
  seed = 42L
)

Arguments

nboot: An integer vector of tiny bootstrap sizes to sweep (the slope and floor are fit over these).
nrow: An integer vector of at least two sample sizes to fit one model per (the nrow_factor is derived from these).
data: A reference data frame with a numeric Conc column to resample (default ssddata::ccme_boron).
seed: A whole number seeding the resampling so the sweep is reproducible.

Value

An ssdsims_cost_calibration object with per-ci_method base/slope/n0 coefficients, a bounded nrow_factor, a fixed_addend, and provenance (cpu, R version, ssdtools version, date, sweep grid).

Details

The sweep is self-contained and dependency-light: it draws data by resampling a reference dataset (ssddata::ccme_boron by default) and times with base system.time(). It takes minutes - not the hours a real scenario costs - because the nboot values are tiny (the slope and floor are estimable from small bootstraps). The one-time research that discovered the model's form (which axes are free, the max(nboot, n0) shape, the non-monotonic nrow factor) is preserved under the change's exploration/ directory; it is not rerun here.

Examples

if (FALSE) { # \dontrun{
calibration <- ssd_calibrate_cost()
calibration
} # }