jnkepler package

Subpackages

• jnkepler.data package
  • Module contents
• jnkepler.jaxttv package
  • Submodules
  • jnkepler.jaxttv.conversion module
    • a2cm_map()
    • astrocentric_to_cm()
    • cm_to_astrocentric()
    • elements_to_xv()
    • j2a_map()
    • jacobi_to_astrocentric()
    • m_to_u()
    • reduce_angle()
    • tic_to_m()
    • tic_to_u()
    • xv_to_elements()
    • xvjac_to_xvacm()
    • xvjac_to_xvcm()
  • jnkepler.jaxttv.findtransit module
    • find_transit_params_all()
    • find_transit_params_fast()
    • find_transit_times_all()
    • find_transit_times_fast()
    • find_transit_times_kepler_all()
  • jnkepler.jaxttv.hermite4 module
    • hermite4_step_map()
    • integrate_xv()
  • jnkepler.jaxttv.infer module
    • get_flat_param_index()
    • make_phase_to_tic_transform()
    • scale_pdic()
    • ttv_default_parameter_bounds()
    • ttv_optim_curve_fit()
    • ttv_optim_least_squares()
    • unscale_pdic()
  • jnkepler.jaxttv.information module
    • information()
    • scale_information()
  • jnkepler.jaxttv.jaxttv module
    • JaxTTV
      • JaxTTV.check_residuals()
      • JaxTTV.check_timing_precision()
      • JaxTTV.get_transit_times_all()
      • JaxTTV.get_transit_times_all_list()
      • JaxTTV.get_transit_times_and_rvs_obs()
      • JaxTTV.get_transit_times_obs()
      • JaxTTV.linear_ephemeris()
      • JaxTTV.plot_model()
      • JaxTTV.sample_means_and_stds()
      • JaxTTV.set_tcobs()
      • JaxTTV.tcall_linear()
      • JaxTTV.transit_time_method
    • Nbody
  • jnkepler.jaxttv.markley module
    • get_E()
  • jnkepler.jaxttv.rv module
    • rv_from_xvjac()
  • jnkepler.jaxttv.symplectic module
    • integrate_xv()
    • kepler_kick_kepler_map()
    • kepler_step_map()
    • kick_kepler_map()
  • jnkepler.jaxttv.ttvfastutils module
    • get_ttvfast_model()
    • get_ttvfast_model_all()
    • get_ttvfast_model_rv()
    • params_for_ttvfast()
  • jnkepler.jaxttv.utils module
    • convert_elements()
    • dict_to_params()
    • elements_to_pdic()
    • em_to_dict()
    • find_nearest_idx()
    • find_nearest_idx_sorted()
    • get_energy_diff()
    • get_energy_diff_jac()
    • initialize_jacobi_xv()
    • params_to_dict()
    • params_to_elements()
  • Module contents
• jnkepler.keplerian package
  • Submodules
  • jnkepler.keplerian.jacobian module
    • det_jkep2D_am()
    • det_jkep2D_pm()
    • det_jkep_am()
    • det_jkep_atau()
    • det_jkep_pm()
    • det_jkep_ptau()
    • slogdet_jkep2D_jax()
    • slogdet_jkep_jax()
  • jnkepler.keplerian.orbit module
    • elements_to_xv()
    • elements_to_xv_scaled()
    • radial_velocity_shape()
    • radial_velocity_shape_multi()
    • tic_to_tau()
    • xv_to_elements()
  • Module contents
• jnkepler.nbodyrv package
  • Submodules
  • jnkepler.nbodyrv.nbodyrv module
    • NbodyRV
      • NbodyRV.get_rv()
  • Module contents
• jnkepler.nbodytransit package
  • Submodules
  • jnkepler.nbodytransit.nbodytransit module
    • NbodyTransit
      • NbodyTransit.get_flux()
      • NbodyTransit.get_flux_and_rv()
      • NbodyTransit.get_xvsky_tc()
      • NbodyTransit.set_lcobs()
    • b_to_cosi()
    • q_to_u()
  • jnkepler.nbodytransit.transit module
    • compute_nbody_flux()
    • compute_nbody_flux_nooverlap()
    • get_xvast_map()
  • Module contents
• jnkepler.tests package
  • Submodules
  • jnkepler.tests.read_testdata module
  • Module contents

Submodules

jnkepler.infer module

jnkepler.infer.fit_t_distribution(y, plot=True, fit_mean=False, save=None, xrange=5)

fit Student’s t distribution to a sample y

Parameters:

• y – 1D array

• plot – if True, plot results

• fit_mean – if True, mean of the distribution is also fitted

Returns:

dictionary with the following keys:

• lndf_loc: mean of log(dof)

• lndf_scale: std of log(dof)

• lnvar_loc: mean of log(variance)

• lnvar_scale: std of log(variance)

• mean_loc: mean of mean (if fitted)

• mean_scale: std of mean (if fitted)

Return type:

dict

jnkepler.infer.optim_svi(numpyro_model, step_size, num_steps, p_initial=None, **kwargs)

optimization using Stochastic Variational Inference (SVI)

Parameters:

• numpyro_model – numpyro model

• step_size – step size for optimization

• num_steps – # of steps for optimization

• p_initial – initial parameter set (dict); if None, use init_to_sample to initialize

Returns:

dictionary containing optimized parameters

Return type:

dict

jnkepler.information module

jnkepler.information.information_from_model_independent_normal(*, model=None, model_args=(), model_kwargs=None, pdic=None, mu_name=None, observed=None, obs_name=None, keys=None, sigma_sd=None, param_space='unconstrained', rng_key=None)

Compute Fisher information matrix for independent Gaussian likelihood directly from a NumPyro model, using (observed - mu(pdic)) / sigma_sd obtained from a NumPyro model.

Parameters:

• model – NumPyro model.

• model_args – static args/kwargs for the model.

• model_kwargs – static args/kwargs for the model.

• pdic – dict of parameter values in constrained space.

• mu_name – deterministic site name for the model mean.

• observed – 1D array of observed values; obs_name is used if not provided.

• obs_name – observed site name.

• keys – list of parameter names to differentiate (order preserved).

• sigma_sd – 1D array of standard deviations (SD) for iid noise.

• param_space – ‘constrained’ or ‘unconstrained’; use ‘unconstrained’ to initialize inverse_mass_matrix.

• rng_key – PRNG key (default = jax.random.PRNGKey(0)).

Returns:

A dictionary containing the Fisher information results and related metadata:

• ”fisher” (jnp.ndarray): The (P, P) Fisher information matrix.

• ”col_slices” (dict[str, slice]): Mapping from each parameter name to its corresponding column range in the Fisher matrix.

• ”col_names” (list[str]): Flattened per-column names, matching the order of columns in the Fisher matrix.

• ”params_unconstrained” (dict[str, jnp.ndarray]): Parameter values in the unconstrained space used for differentiation.

Return type:

dict

jnkepler.jnkepler_version module

Module contents