add_estimation_step#

pharmpy.modeling.add_estimation_step(model, method, idx=None, interaction=False, parameter_uncertainty_method=None, evaluation=False, maximum_evaluations=None, laplace=False, isample=None, niter=None, auto=None, keep_every_nth_iter=None, residuals=(), predictions=(), solver=None, solver_rtol=None, solver_atol=None, tool_options={}, derivatives=(), individual_eta_samples=False)[source]#

Add estimation step

Adds estimation step for a model in a given index. Methods currently supported are:

FO, FOCE, ITS, LAPLACE, IMPMAP, IMP, SAEM

Parameters:

  • model (Model) – Pharmpy model

  • method (MethodType) – estimation method to change to

  • idx (int) – index of estimation step (starting from 0), default is None (adds step at the end)

  • interaction (bool) – See EstimationStep for more information on options

  • parameter_uncertainty_method (Optional[str]) – See above

  • evaluation (bool) – See above

  • maximum_evaluations (Optional[int]) – See above

  • laplace (bool) – See above

  • isample (Optional[int]) – See above

  • niter (Optional[int]) – See above

  • auto (Optional[bool]) – See above

  • keep_every_nth_iter (Optional[int]) – See above

  • residuals (Sequence[str]) – See above

  • predictions (Sequence[str]) – See above

  • solver (Optional[str]) – See above

  • solver_rtol (Optional[int]) – See above

  • solver_atol (Optional[int]) – See above

  • tool_options (Mapping[str, Any]) – See above

  • derivatives (Sequence[Sequence[Expr]]) – See above

  • individual_eta_samples (bool) – See above

Returns:

Model – Pharmpy model object

Examples

>>> from pharmpy.modeling import *
>>> model = load_example_model("pheno")
>>> opts = {'NITER': 1000, 'ISAMPLE': 100}
>>> model = add_estimation_step(model, 'IMP', tool_options=opts)
>>> ests = model.execution_steps
>>> len(ests)
2
>>> ests[1]
EstimationStep('IMP', interaction=False, parameter_uncertainty_method=None, ..., tool_options={'NITER': 1000,...

See also

set_estimation_step, remove_estimation_step, append_estimation_step_options, add_parameter_uncertainty_step, remove_parameter_uncertainty_step, set_evaluation_step

On this page