lightcurvelynx.models.bayesn

Classes

BayesnModel

A bayesian model for supernova type Ia

Module Contents

class BayesnModel(theta=None, Av=None, Rv=None, t0=0.0, Amplitude=1.0, M20_path_or_url='https://github.com/bayesn/bayesn-model-files/raw/refs/heads/main/BAYESN.M20/', W0_filename='W0.txt', W1_filename='W1.txt', l_knots_filename='l_knots.txt', tau_knots_filename='tau_knots.txt', hsiao_path_or_url='https://github.com/bayesn/bayesn/raw/refs/heads/main/bayesn/data/hsiao.h5', **kwargs)[source]

Bases: lightcurvelynx.models.physical_model.SEDModel, citation_compass.CiteClass

A bayesian model for supernova type Ia

The model is defined in (Mandel et al 2022) as:

flux(time, wave) = H_grid * 10 ** (-0.4 * W_grid) * 10 ** (-0.4 * (distmod _ m_abs))

This class is based on the bayesian implementation at: https://github.com/bayesn/bayesn/blob/main/bayesn/bayesn_model.py

Parameterized values include:

  • ra - The object’s right ascension in degrees. [from BasePhysicalModel]

  • dec - The object’s declination in degrees. [from BasePhysicalModel]

  • redshift - The object’s redshift. [from BasePhysicalModel]

  • t0 - The t0 of the zero phase, date. [from BasePhysicalModel]

  • Amplitude - The fixed normalisation factor for distance modulus. [from Amplitude class]

  • theta - The bayeSN theta parameter.

  • Av - The bayeSN Av parameter.

  • Rv - The bayeSN Rv parameter.

References

  • BayeSN: Mandel S., 2020 - https://arxiv.org/pdf/2008.07538

  • M20 model: Mandel et al. 2022 (MNRAS 510, 3, 3939-3966)

  • T21 model: Thorp et al. 2021 (MNRAS 508, 3, 4310-4331) - currently unused

  • W22 model: Ward et al. 2023 (ApJ 956, 2, 111) - currently unused

  • Hsiao spectral template: Hsiao E. Y., 2009 - https://arxiv.org/abs/1503.02293

_hsiao_phase

The phase for hsiao template.

Type:

numpy.ndarray

_hsiao_wave

The wavelengths for hsiao template.

Type:

numpy.ndarray

_hsiao_flux

The baseline mean intrinsic SED provided by hsiao template.

Type:

numpy.ndarray

_W0_[source]

The global W0 matrix.

Type:

numpy.ndarray

_W1_[source]

The global W1 matrix.

Type:

numpy.ndarray

_l_knots_[source]

The interpolation knots for wavelengths.

Type:

numpy.ndarray

_tau_knots_[source]

The interpolation knots for phase.

Type:

numpy.ndarray

Parameters:

  • theta (parameter) – The bayeSN theta parameter.

  • Av (parameter) – The bayeSN Av parameter. Set of host extinction values for each SN.

  • Rv (parameter) – The bayeSN Rv parameter. Rv value for host extinction.

  • Amplitude (parameter) – The distance modulus (m - M) Normalized to have a absolute magnitude of -19.5

  • _M20_model_path (str) – The path for the M20 model file directory from the bayesian model Default: “bayesn-model-files/BAYESN.M20”

  • W0_filename (str) – The file name of the W0 matrix. Default: “W0.txt”

  • W1_filename (str) – The file name of the W1 matrix. Default: “W1.txt”

  • l_knots_filename – The file name of the knot values of wavelegnths for interpolation Default: “l_knots.txt”

  • tau_knots_filename (str) – The file name of the knot values of times for interpolation Default: “tau_knots.txt”

  • hsiao_path_or_url (str) – The path for the hsiao model template file directory. Default: “bayesn-model-files/hsiao.h5”

  • **kwargs (dict, optional) – Any additional keyword arguments.

minphase(**kwargs)[source]

Get the minimum supported rest-frame phase of the model in days.

Parameters:

**kwargs (dict) – Additional keyword arguments, not used in this method.

Returns:

minphase – The minimum phase of the model (in days) or None if the model does not have a defined minimum phase.

Return type:

float or None

maxphase(**kwargs)[source]

Get the minimum supported rest-frame phase of the model in days.

Parameters:

**kwargs (dict) – Additional keyword arguments, not used in this method.

Returns:

minphase – The minimum phase of the model (in days) or None if the model does not have a defined minimum phase.

Return type:

float or None

compute_invkd(x)[source]

Compute the operator matrix K^{-1}D to get second derivatives for natural cubic spline.

Parameters:

x ((n,) array_like) – Knot positions (non-uniform, strictly increasing).

Returns:

invKD – Matrix such that M = invKD @ y gives second derivatives of y.

Return type:

(n, n) ndarray

natural_cubic_spline_basis_matrix_from_invkd(x, xq_array, invKD)[source]

Compute basis matrix J for multiple query points using precomputed second derivative matrix.

Parameters:

  • x ((n,) array_like) – Knot positions.

  • xq_array ((m,) array_like) – Query points.

  • invKD ((n, n) array_like) – Precomputed matrix such that second_derivatives = invKD @ y.

Returns:

J – Basis matrix. Each row J[i, :] is the spline basis vector for xq_array[i].

Return type:

(m, n) ndarray

compute_second_derivatives_1d(x, y)[source]

Compute natural cubic spline second derivatives (M) for 1D input.

Parameters:

  • x ((n,) array_like) – Knot positions.

  • y ((n,) array_like) – Values at knots.

Returns:

M – Second derivatives at knots.

Return type:

(n,) ndarray

compute_2d_second_derivatives(x, y, z)[source]

Compute second derivatives in both x and y directions for 2D natural cubic spline.

Parameters:

  • x ((n,) array_like) – Knot positions in x.

  • y ((m,) array_like) – Knot positions in y.

  • z ((n, m) array_like) – Function values on 2D grid.

Returns:

  • Mx ((n, m) ndarray) – Second derivatives with respect to x.

  • My ((n, m) ndarray) – Second derivatives with respect to y.

evaluate_natural_spline_2d_vectorized(x, y, z, Mx, My, xq, yq)[source]

Vectorized 2D natural cubic spline evaluation using second derivatives.

Parameters:

  • x ((n,) array_like) – Knot positions in x.

  • y ((m,) array_like) – Knot positions in y.

  • z ((n, m) array_like) – Function values on 2D grid.

  • Mx ((n, m) array_like) – Second derivatives w.r.t. x.

  • My ((n, m) array_like) – Second derivatives w.r.t. y.

  • xq ((p,) array_like) – Query positions in x.

  • yq ((q,) array_like) – Query positions in y.

Returns:

Zq – Interpolated 2D surface values.

Return type:

(p, q) ndarray

evaluate_2d_cubic_spline(x, y, z, xq, yq)[source]

Evaluate 2D natural cubic spline at given query points with auto second derivative computation.

Parameters:

  • x ((n,) array_like) – Knot positions in x direction.

  • y ((m,) array_like) – Knot positions in y direction.

  • z ((n, m) array_like) – Function values at grid points.

  • xq ((p,) array_like) – Query points in x.

  • yq ((q,) array_like) – Query points in y.

Returns:

Zq – Interpolated values at query grid.

Return type:

(p, q) ndarray

compute_sed(times, wavelengths, graph_state, **kwargs)[source]

Draw effect-free observations for this object.

Parameters:

  • times (numpy.ndarray) – A length T array of rest frame timestamps in MJD.

  • wavelengths (numpy.ndarray, optional) – A length N array of rest frame wavelengths (in angstroms).

  • graph_state (GraphState) – An object mapping graph parameters to their values.

  • **kwargs (dict, optional) – Any additional keyword arguments.

Returns:

flux_density – A length T x N matrix of SED values (in nJy).

Return type:

numpy.ndarray