#!/usr/bin/env python
# -*- coding: utf-8 -*-
# Copyright (c) 2017-2024, Cabral, Juan
# Copyright (c) 2025, QuatroPe; Clariá, Felipe
# License: MIT
# Full Text:
# https://github.com/quatrope/feets/blob/master/LICENSE
# =============================================================================
# DOC
# =============================================================================
"""Villar fit extractor."""
# =============================================================================
# IMPORTS
# =============================================================================
from light_curve import VillarFit as _VillarFit
from .light_curve_extractor import LightCurveExtractor
from ..libs import doctools
# =============================================================================
# EXTRACTOR CLASS
# =============================================================================
[docs]
class VillarFit(LightCurveExtractor):
r"""Villar function fit.
Seven fit parameters and goodness of fit (reduced :math:`\chi^2`) of
the Villar function developed for supernovae classification:
.. math::
f(t) = c + \frac{A}{1 + \exp{\frac{-(t-t_0)}{\tau_\mathrm{rise}}}}
\left\{
\begin{array}{ll}
1 - \frac{\nu (t - t_0)}{\gamma}, &t < t_0 + \gamma \\
(1 - \nu) \exp{\frac{-(t-t_0-\gamma)}{\tau_\mathrm{fall} }},
&t \geq t_0 + \gamma
\end{array}
\right.
where :math:`A, \gamma, \tau_\mathrm{rise}, \tau_\mathrm{fall} > 0`,
:math:`\nu \in [0; 1)`.
Here we introduce a new dimensionless parameter :math:`\nu` instead of
the plateau slope :math:`\beta` from the original paper:
.. math::
\nu \equiv -\beta \gamma / A
Note, that the Villar function is developed to be used with fluxes,
not magnitudes.
**VillarFit_Amplitude** (:math:`A`)
Half amplitude of the Villar function
**VillarFit_Baseline** (:math:`c`)
Baseline of the Villar function
**VillarFit_ReferenceTime** (:math:`t_0`)
Reference time of the Villar function
**VillarFit_RiseTime** (:math:`\tau_\mathrm{rise}`)
Rise time of the Villar function
**VillarFit_FallTime** (:math:`\tau_\mathrm{fall}`)
Decline time of the Villar function
**VillarFit_PlateauRelAmplitude** (:math:`\nu = -\beta \gamma / A`)
Relative plateau amplitude of the Villar function
**VillarFit_PlateauDuration** (:math:`\gamma`)
Plateau duration of the Villar function
**VillarFit_ReducedChi2** (reduced :math:`\chi^2`)
Villar fit quality
Parameters
----------
algorithm : str
Non-linear least-square algorithm, supported values are:
'mcmc', 'ceres', 'mcmc-ceres', 'lmsder', 'mcmc-lmsder'.
mcmc_niter : int, default=128
Number of MCMC iterations.
ceres_niter : int, default=10
Number of Ceres iterations.
ceres_loss_reg : float, optional
Ceres loss regularization, default is to use square norm as is, if set
to a number, the loss function is regularized to discriminate outlier
residuals larger than this value.
Default is None which means no regularization.
lmsder_niter : int, default=10
Number of LMSDER iterations.
init : list or None, optional
Initial conditions, must be `None` or a `list` of `float`s or `None`s.
The length of the list must be 7, `None` values will be replaced
with some default values. It is supported by MCMC only
bounds : list of tuples or None, optional
Boundary conditions, must be `None` or a `list` of `tuple`s of `float`s
or `None`s. The length of the list must be 7, boundary conditions must
include initial conditions, `None` values will be replaced with some
broad defaults. It is supported by MCMC only
ln_prior : str or list of ln_prior.LnPrior1D or None, optional
Prior for MCMC, None means no prior. It is specified by a string
literal or a list of 7 `light_curve.ln_prior.LnPrior1D` objects, see
`light_curve.ln_prior` submodule for corresponding functions. Available
string literals are:
- 'no': no prior
- 'hosseinzadeh2020': prior adopted from Hosseinzadeh et al. 2020, it
assumes that `t` is in days
transform : str or bool or None, optional
If `False` or `None` (default) output is not transformed. If `True` output
is transformed as following:
- Half-amplitude A is transformed as :math:`zp - 2.5 lg(2*A)`,
:math:`zp = 8.9`, so that the amplitude is assumed to be the object
peak flux in Jy.
- baseline flux is normalised by :math:`A: baseline -> baseline / A`
- reference time is removed
- goodness of fit is transformed as :math:`ln(reduced chi^2 + 1)` to
reduce its spread
- other parameters are not transformed
"""
features = [
"VillarFit_Amplitude",
"VillarFit_Baseline",
"VillarFit_ReferenceTime",
"VillarFit_RiseTime",
"VillarFit_FallTime",
"VillarFit_PlateauRelAmplitude",
"VillarFit_PlateauDuration",
"VillarFit_ReducedChi2",
]
def __init__(
self,
algorithm="mcmc",
mcmc_niter=128,
lmsder_niter=10,
ceres_niter=10,
ceres_loss_reg=None,
init=None,
bounds=None,
ln_prior=None,
transform=None,
):
self.algorithm = algorithm
self.mcmc_niter = mcmc_niter
self.lmsder_niter = lmsder_niter
self.ceres_niter = ceres_niter
self.ceres_loss_reg = ceres_loss_reg
self.init = init
self.bounds = bounds
self.ln_prior = ln_prior
self.transform = transform
self._extract = _VillarFit(**self.params)
