SamplingParameter#

jeanspy.parameters.SamplingParameter

See methods and properties for individual lookup pages, or the alphabetical API dictionary to search all classes. The full class contract and existing member anchors are retained below.

class jeanspy.parameters.SamplingParameter(sample_name, param_name, transform='identity')[source]#

Bases: object

Map one sampled coordinate to a physical model parameter.

Parameters:

  • sample_name (str) – Coordinate name, matching one row in the finite-prior table.

  • param_name (str) – Physical model parameter receiving the transformed value.

  • transform (str, optional) – identity (default), pow10 (10**x), one_minus_pow10 (1-10**x), or arccos (radians). Names never select a transform.

Notes

Prior bounds and photometric priors are evaluated in sampled coordinates. No Jacobian is added: a uniform bound on x with pow10 defines a log-uniform physical parameter. Supply specifications in prior-table order. Invalid names or transform choices raise ValueError at construction. Conversion uses NumPy and is not a JAX tracing interface.

Examples

>>> SamplingParameter("log_radius", "re_pc", "pow10").to_physical(2.)
np.float64(100.0)
sample_name: str#
param_name: str#
transform: str = 'identity'#
to_physical(value)[source]#

Convert a scalar/array, preserving shape; invalid values yield NaN/inf.

Physical units are those of param_name; arccos returns radians. The consuming likelihood is responsible for its physical-domain check.