openghg_inversions.models.priors#

Create reusable scalar or array-valued PyMC prior variables.

parse_prior registers continuous distributions on the active model; lognormal_mu_sigma converts requested moments, including arrays. Optional lognormal reparameterization exposes <name>_latent and keeps <name> as the user-facing deterministic variable.

openghg_inversions.models.priors.lognormal_mu_sigma(mean: float | ndarray, stdev: float | ndarray) → tuple[float | ndarray, float | ndarray]#

Convert lognormal mean and stdev into PyMC’s mu and sigma.

Parameters:

  • mean – Requested scalar or array-valued mean of the lognormal distribution.

  • stdev – Requested scalar or array-valued standard deviation of the lognormal distribution.

Returns:

A (mu, sigma) tuple suitable for pm.Lognormal.

openghg_inversions.models.priors.parse_prior(name: str, prior_params: dict[str, Any], **kwargs) → TensorVariable#

Create a continuous PyMC prior from a prior-parameter dictionary.

Parameters:

  • name – Name of the user-facing PyMC variable to create.

  • prior_params – Prior specification including pdf and any distribution parameters accepted by the chosen PyMC distribution.

  • **kwargs – Additional keyword arguments forwarded to the created PyMC variable, such as dims.

Returns:

The created PyMC random variable or deterministic transform.

Raises:

ValueError – If prior_params["pdf"] does not name a supported PyMC continuous distribution.

This helper must be called inside an active pm.Model context because it registers the created variable with the current model.