Skip to contents

Benchmark wrapper: Pilar 6 – bootstrap-ensembled quantum KRR

Source: R/benchmark_wosis.R
benchmark_fit_p6_quantum.Rd

PCA-reduces the covariate matrix to n_pcs components rescaled to [-pi, pi], then trains n_boot quantum-kernel ridge regressors on bootstrap resamples and aggregates their predictions into a predictive posterior.

Usage

benchmark_fit_p6_quantum(
  train,
  test,
  cov_cols,
  n_pcs = 6L,
  reps = 2L,
  lambda = 0.5,
  n_boot = 20L,
  seed = 1L,
  calibrate = TRUE
)

Arguments

train, test

Data frames with soc + cov_cols.

cov_cols

Character vector of covariate column names.

n_pcs

Integer; number of PCs (= qubits). Default 6L.

reps

Integer; ZZFeatureMap repetitions. Default 2L.

lambda

Ridge regulariser. Default 0.5.

n_boot

Integer; number of bootstrap KRR fits. Default 20L.

seed

Optional RNG seed.

calibrate

Logical (default TRUE). When TRUE, residual noise from a full-data quantum-KRR fit is injected into every posterior sample (see Pilar-1 wrapper for the rationale).

Value

An edaphos_posterior with method = "ensemble", query_type = "map".