petab.sampling

Functions related to parameter sampling

Functions

sample_from_prior(prior, n_starts)

Creates samples for one parameter based on prior

sample_parameter_startpoints(parameter_df[, ...])

Create numpy.array with starting points for an optimization
petab.sampling.sample_from_prior(prior: Tuple[str, list, str, list], n_starts: int) → array[source]

Creates samples for one parameter based on prior

Parameters:

  • prior – A tuple as obtained from petab.parameter.get_priors_from_df()

  • n_starts – Number of samples

Returns:

Array with sampled values

petab.sampling.sample_parameter_startpoints(parameter_df: DataFrame, n_starts: int = 100, seed: int | None = None, parameter_ids: Sequence[str] | None = None) → array[source]

Create numpy.array with starting points for an optimization

Parameters:

  • parameter_df – PEtab parameter DataFrame

  • n_starts – Number of points to be sampled

  • seed – Random number generator seed (see numpy.random.seed())

  • parameter_ids – A sequence of parameter IDs for which to sample starting points. For subsetting or reordering the parameters. Defaults to all estimated parameters.

Returns:

Array of sampled starting points with dimensions n_startpoints x n_optimization_parameters