Parameters

class pypomp.PompParameters(theta: Mapping[str, float] | Sequence[Mapping[str, float]] | PompParameters | DataArray | Dataset | None, logLik: ndarray | None = None, estimation_scale: bool = False)

Bases: ParameterSet

Parameter set for standard POMP models.

Internally wraps an xarray.Dataset with two variables, shared (dims ("theta_idx", "parameter")) and an empty unit_specific. A standard POMP has no units, so all parameters live in shared. This is the same representation used by PanelParameters, allowing both to share the replicate/log-likelihood machinery defined in the base class.

Parameters:

• theta (mapping or sequence of mapping or PompParameters or xr.DataArray or None) –

  Parameters for the model. Accepts:

  • A single dictionary: dict[str, float]

  • A sequence of dictionaries: list[dict[str, float]]

  • An existing PompParameters object

  • An xarray.DataArray (1D over parameters, 2D over (theta_idx, parameter), or 3D with a singleton unit)

• logLik (np.ndarray, optional) – Log-likelihood values associated with each parameter set.

• estimation_scale (bool, optional) – Whether the parameters are on the estimation scale. Defaults to False.

Methods

logLik	Get or set the log-likelihoods for each parameter set (theta_idx).
to_jax_array([param_names])	Convert parameter values to a JAX array matching param_names.
num_replicates()	Returns the number of parameter sets/replicates.
num_params()	Return the number of canonical parameters.
subset(indices)	Return a new parameter set with only the selected replicates.
get_param_names()	Return the parameter names in their original (insertion) order.
params(-> list[dict[str, ...)	Get the parameter values in this set.
set_params(value)	Overwrite parameter values.
transformed(par_trans[, direction])	Transform parameters between natural and estimation scales.
pruned([n, refill])	Return a new parameter set with the top n replicates by log-likelihood.
merge(*param_objs)	Merge replicates from multiple parameter sets of this type.

class pypomp.PanelParameters(theta: Mapping[str, DataFrame | None] | Sequence[Mapping[str, DataFrame | None]] | PanelParameters | Dataset | None, logLik_unit: ndarray | None = None, estimation_scale: bool = False)

Bases: ParameterSet

Parameter set for panel POMP models.

Manages parameters partitioned into shared and unit-specific parameters.

Parameters:

• theta (mapping or sequence of mapping or PanelParameters or xr.Dataset or None) –

  Parameters for the panel model. Accepts:

  • A dictionary with keys "shared" and "unit_specific" mapping to dataframes.

  • A sequence of such dictionaries.

  • An existing PanelParameters object.

  • An xarray.Dataset containing shared and unit-specific parameter variables.

• logLik_unit (np.ndarray, optional) – Unit-specific log-likelihoods of shape (n_reps, n_units).

• estimation_scale (bool, optional) – Whether the parameters are on the estimation scale. Defaults to False.

Methods

logLik	Get the overall log-likelihood for each parameter set (theta_idx).
logLik_unit	Get or set the unit-specific log-likelihoods for each parameter set (theta_idx).
params(-> list[dict[str, ...)	Get the parameter values in this set.
set_params(value)	Overwrite parameter values.
num_replicates()	Returns the number of parameter sets/replicates.
get_param_names()	Return the list of parameter names contained in this set.
get_shared_param_names()	Get the list of shared parameter names.
get_unit_param_names()	Get the list of unit-specific parameter names.
get_unit_names()	Get the list of unit names.
subset(indices)	Return a new parameter set with only the selected replicates.
to_jax_array([param_names, unit_names])	Convert parameter values to a JAX array.
transformed(par_trans[, direction])	Transform parameters between natural and estimation scales.
pruned([n, refill])	Return a new parameter set with the top n replicates by log-likelihood.
mixed_and_matched()	Sort parameters independently and cross-combine them.
merge(*param_objs)	Merge replicates from multiple parameter sets of this type.