Model¶

class vega.model.Model(corr_item, fiducial, scale_params, data=None)[source]¶

Class for computing Lyman-alpha forest correlation function models.

Parameters:	corr_item (CorrelationItem) – Item object with the component config fiducial (dict) – fiducial config scale_params (ScaleParameters) – ScaleParameters object data (Data, optional) – data object corresponding to the cf component, by default None

compute(pars, pk_full, pk_smooth)[source]¶

Compute correlation function model using the peak/smooth (wiggles/no-wiggles) decomposition.

Parameters:	pars (dict) – Computation parameters pk_full (1D Array) – Full fiducial linear power spectrum pk_smooth (1D Array) – Smooth component of the fiducial linear power spectrum
Returns:	Full correlation function
Return type:	1D Array

compute_direct(pars, pk_full)[source]¶

Compute full correlation function model directly from the full power spectrum.

Parameters:	pars (dict) – Computation parameters pk_full (1D Array) – Full fiducial linear power spectrum
Returns:	Full correlation function
Return type:	1D Array

static init_broadband(bb_input, cf_name, bin_size_rp_data, bin_size_rp_model)[source]¶

Read the broadband config and initialize what we need.

Parameters:	bb_input (ConfigParser) – broadband section from the config file cf_name (string) – Name of corr item bin_size_rp_data (float) – Size of data r parallel bins bin_size_rp_model (float) – Size of model r parallel bins
Returns:	list with configs of broadband terms
Return type:	list

class vega.correlation_func.CorrelationFunction(config, fiducial, coords_grid, scale_params, tracer1, tracer2, bb_config=None, metal_corr=False)[source]¶

Correlation function computation and handling.

# ! Slow operations should be kept in init as that is only called once

# ! Compute is called many times and should be fast

Extensions should have their separate method of the form ‘compute_extension’ that can be called from outside

Parameters:

config (ConfigParser) – model section of config file
fiducial (dict) – fiducial config
coords_grid (dict) – Dictionary with coordinate grid - r, mu, z
scale_params (ScaleParameters) – ScaleParameters object
tracer1 (dict) – Config of tracer 1
tracer2 (dict) – Config of tracer 2
bb_config (list, optional) – list with configs of broadband terms, by default None
metal_corr (bool, optional) – Whether this is a metal correlation, by default False

broadband(bb_term, params)[source]¶

Compute broadband term.

Calculates a power-law broadband in r and mu or rp,rt.

Parameters:	bb_term (dict) – broadband term config params (dict) – Computation parameters
Returns:	Output broadband
Return type:	1d Array

broadband_sky(bb_term, params)[source]¶

Compute sky broadband term.

Calculates a Gaussian broadband in rp,rt for the sky residuals.

Parameters:	bb_term (dict) – broadband term config params (dict) – Computation parameters
Returns:	Output broadband
Return type:	1d Array

compute(pk, pk_lin, PktoXi_obj, params)[source]¶

Compute correlation function for input P(k).

Parameters:	pk (ND Array) – Input power spectrum pk_lin (1D Array) – Linear isotropic power spectrum PktoXi_obj (vega.PktoXi) – An instance of the transform object used to turn Pk into Xi params (dict) – Computation parameters
Returns:	Output correlation function
Return type:	1D Array

compute_bias_evol(params)[source]¶

Compute bias evolution for the correlation function.

Parameters:	params (dict) – Computation parameters
Returns:	Bias evolution for tracer
Return type:	ND Array

compute_broadband(params, pos_type)[source]¶

Compute the broadband terms for one position (pre-distortion/post-distortion) and one type (multiplicative/additive).

Parameters:	params (dict) – Computation parameters pos_type (string) – String with position and type, must be one of: ‘pre-mul’ or ‘pre-add’ or ‘post-mul’ or ‘post-add’
Returns:	Output broadband
Return type:	1d Array

compute_core(pk, PktoXi_obj, params)[source]¶

Compute the core of the correlation function.

This does the Hankel transform of the input P(k), sums the necessary multipoles and rescales the coordinates

Parameters:	pk (ND Array) – Input power spectrum PktoXi_obj (vega.PktoXi) – An instance of the transform object used to turn Pk into Xi params (dict) – Computation parameters
Returns:	Output correlation function
Return type:	1D Array

compute_desi_instrumental_systematics(params, bin_size_rp)[source]¶

Compute DESI instrumental systematics model TODO add link to Satya’s paper describing this

Parameters:	params (dict) – Computation parameters bin_size_rp (float) – Bin size along the line-of-sight
Returns:	Output correction
Return type:	1D Array

compute_growth(z_grid=None, z_fid=None, Omega_m=None, Omega_de=None)[source]¶

Compute growth factor.

Implements eq. 7.77 from S. Dodelson’s Modern Cosmology book.

Returns:	Growth factor
Return type:	ND Array

compute_qso_radiation(params)[source]¶

Model the contribution of QSO radiation to the cross (the transverse proximity effect)

Parameters:	params (dict) – Computation parameters
Returns:	Xi QSO radiation model
Return type:	1D

compute_xi_asymmetry(pk, PktoXi_obj, params)[source]¶

Calculate the cross-correlation contribution from standard asymmetry (Bonvin et al. 2014).

Parameters:	pk (ND Array) – Input power spectrum PktoXi_obj (vega.PktoXi) – An instance of the transform object used to turn Pk into Xi params (dict) – Computation parameters
Returns:	Output xi asymmetry
Return type:	1D Array

compute_xi_relativistic(pk, PktoXi_obj, params)[source]¶

Calculate the cross-correlation contribution from relativistic effects (Bonvin et al. 2014).

Parameters:	pk (ND Array) – Input power spectrum PktoXi_obj (vega.PktoXi) – An instance of the transform object used to turn Pk into Xi params (dict) – Computation parameters
Returns:	Output xi relativistic
Return type:	1D Array

class vega.power_spectrum.PowerSpectrum(config, fiducial, tracer1, tracer2, dataset_name=None)[source]¶

Power Spectrum computation and handling.

# ! Slow operations should be kept in init as that is only called once

# ! Compute is called many times and should be fast

Extensions should have their separate method of the form ‘compute_extension’ that can be called from outside

Parameters:	config (dict) – pk config object fiducial (dict) – fiducial config tracer1 (dict) – Config of tracer 1 tracer2 (dict) – Config of tracer 1 dataset_name (string) – Name of dataset, by default None

compute(pk_lin, params, fast_metals=False)[source]¶

Computes a power spectrum for the tracers using the input linear P(k) and parameters.

Parameters:	pk_lin (1D array) – Linear Power Spectrum params (dict) – Computation parameters
Returns:	Power spectrum
Return type:	ND Array

compute_Gk(params)[source]¶

Model the effect of binning of the cf.

Parameters:	params (dict) – Computation parameters
Returns:	G(k)
Return type:	ND Array

compute_bias_beta_hcd(bias, beta, params)[source]¶

Compute effective biases that include HCD modeling.

Parameters:	bias (float) – Bias for tracer beta (float) – Beta for tracer params (dict) – Computation parameters
Returns:	Effective bias and beta
Return type:	(float, float)

compute_bias_beta_uv(bias, beta, params)[source]¶

Compute effective biases that include UV modeling.

Parameters:	bias (float) – Bias for tracer beta (float) – Beta for tracer params (dict) – Computation parameters
Returns:	Effective bias and beta
Return type:	(float, float)

compute_dnl_arinyo(params)[source]¶

Non linear term from Arinyo et al 2015.

Parameters:	params (dict) – Computation parameters
Returns:	D_NL factor
Return type:	ND Array

compute_dnl_mcdonald()[source]¶

Non linear term from McDonald 2003.

Returns:	D_NL factor
Return type:	ND Array

compute_fullshape_exp_smoothing(params)[source]¶

Compute a Gaussian and exp smoothing for the full correlation function (usefull for london_mocks_v6.0).

Parameters:	params (dict) – Computation parameters
Returns:	Smoothing factor
Return type:	ND Array

compute_fullshape_gauss_smoothing(params)[source]¶

Compute a Gaussian smoothing for the full correlation function.

Parameters:	params (dict) – Computation parameters
Returns:	Smoothing factor
Return type:	ND Array

compute_kaiser(bias1, beta1, bias2, beta2, fast_metals=False)[source]¶

Compute Kaiser model.

Parameters:	bias1 (float) – Bias for tracer 1 beta1 (float) – Beta for tracer 1 bias2 (float) – Bias for tracer 2 beta2 (float) – Beta for tracer 2
Returns:	Kaiser term
Return type:	ND Array

compute_peak_nl(params)[source]¶

Compute the non-linear gaussian correction for the peak component.

Parameters:	params (dict) – Computation parameters
Returns:	Smoothing factor for the peak
Return type:	ND Array

compute_velocity_dispersion_gauss(params)[source]¶

Compute a gaussian smoothing factor to model velocity dispersion.

Parameters:	params (dict) – Computation parameters
Returns:	Smoothing factor
Return type:	ND Array

compute_velocity_dispersion_lorentz(params)[source]¶

Compute a lorentzian smoothing factor to model velocity dispersion.

Parameters:	params (dict) – Computation parameters
Returns:	Smoothing factor
Return type:	ND Array