__call__

jeanspy.sampler_numpyro.AxisymmetricJeansLikelihoodModel.__call__

Class and construction: AxisymmetricJeansLikelihoodModel.

AxisymmetricJeansLikelihoodModel.__call__(x_pc, y_pc, vlos_kms, e_vlos_kms)

Sample parameters and register the axisymmetric velocity likelihood.

Signed sky coordinates x_pc, y_pc (pc), velocities and nonnegative errors (km/s) must be matching, nonempty 1-D arrays. The projected center is allowed. This NumPyro model returns None and creates parameter, validity-factor and observed-velocity sites. Nonfinite data or variances outside sigma2_bounds contribute minus infinity; incompatible array shapes raise ValueError. Measurement errors are added in quadrature to admissible intrinsic variances. Run through a NumPyro inference or handler context; gradients apply within the differentiable, physically admissible model domain.