Skip to content

API Reference

Core Classes

MultipoleEmulators

The main class for computing galaxy power spectrum multipoles.

Methods: - get_Pl(cosmo_params, bias_params, D): Compute power spectrum multipole - P11: Access P11 component emulator - Ploop: Access Ploop component emulator - Pct: Access Pct component emulator

MultipoleNoiseEmulator

Extension of MultipoleEmulators with stochastic noise terms.

Methods: - get_Pl(cosmo_params, bias_params, D, stoch_params): Compute multipole with noise

Cosmology Classes

W0WaCDMCosmology

w₀wₐCDM cosmology with massive neutrinos support.

Parameters: - ln10As: Log amplitude of primordial power spectrum - ns: Spectral index - h: Hubble parameter - omega_b: Baryon density - omega_c: CDM density - m_nu: Neutrino mass [eV] - w0: Dark energy equation of state - wa: Dark energy equation of state evolution

Methods: - D_z(z): Growth factor at redshift z - H_z(z): Hubble parameter at redshift z - comoving_distance(z): Comoving distance to redshift z

Loading Functions

load_multipole_emulator

Load a multipole emulator from disk.

emulator = jaxeffort.load_multipole_emulator(path)

Parameters: - path: Path to emulator directory

Returns: - MultipoleEmulators instance

force_update

Force update of cached emulator data.

jaxeffort.force_update()

Downloads latest emulator data from Zenodo and updates local cache.

Pre-trained Emulators

The following pre-trained emulators are available through jaxeffort.trained_emulators:

pybird_mnuw0wacdm

Emulator trained on PyBird calculations with w₀wₐCDM cosmology and massive neutrinos.

  • Multipoles available: 0 (monopole), 2 (quadrupole), 4 (hexadecapole)
  • Redshift range: 0.5 - 2.0
  • k range: 0.005 - 0.3 h/Mpc
  • Parameters:
  • Cosmological: z, ln10As, ns, H0, ombh2, omch2, Mnu, w0, wa
  • Bias: b1, b2, b3, b4, b5, b6, b7, f

Example usage: 

P0 = jaxeffort.trained_emulators["pybird_mnuw0wacdm"]["0"]
P2 = jaxeffort.trained_emulators["pybird_mnuw0wacdm"]["2"]
P4 = jaxeffort.trained_emulators["pybird_mnuw0wacdm"]["4"]

Utility Functions

get_stoch_terms

Compute stochastic EFT terms.

stoch = jaxeffort.get_stoch_terms(k, ceps0, ceps1, ceps2)

Parameters: - k: Wavenumber array - ceps0: Constant stochastic term - ceps1: Linear stochastic term coefficient - ceps2: Quadratic stochastic term coefficient

Returns: - Array of stochastic contributions