survivalgpu.simulate_for_experiment

survivalgpu.simulate_for_experiment(n_patients, max_time, HR_target, scenario_name, seed=None)

Simulates a dataset with a single WCE “dose” covariate, for benchmarking/validation experiments.

Builds a single WCECovariate named “dose”, with possible values [1, 1.5, 2, 2.5, 3], the given scenario_name exposure-effect shape, and target hazard ratio HR_target, then calls simulate_dataset with it.

Parameters:
  • n_patients (int) – the number of patients to simulate.

  • max_time (int) – the maximum follow-up time.

  • HR_target (float) – the target hazard ratio for the “dose” covariate.

  • scenario_name (str) – name of the exposure-effect scenario (see get_scenario).

  • seed (int, optional) – random seed, for reproducibility. Defaults to None.

Returns:

pandas.DataFrame –

long-format dataset, as returned by simulate_dataset

(with compress=False).