to_inferencedata#

exo_skryer.help_io.to_inferencedata(samples_dict: Dict[str, ndarray], cfg, include_fixed: bool = False) DataTree[source]#

Convert {param: array} mapping into ArviZ InferenceData.

Parameters:
samples_dictdict

Mapping name -> samples. Each value can be shaped (draws,) or (chains, draws). Sampler output after JAX → NumPy conversion.

cfgSimpleNamespace

Parsed YAML configuration. Expects cfg.params as a list of parameter objects with at least attributes: name, dist, and optionally init / value.

include_fixedbool, default False

If True, include delta/fixed parameters in the posterior group. If False, skip them.

Returns:
az.InferenceData

ArviZ InferenceData with a posterior group.