pint.models.parameter.funcParameter

class pint.models.parameter.funcParameter(name=None, description=None, func=None, params=None, units=None, inpar=False, long_double=False, unit_scale=False, scale_factor=None, scale_threshold=None, aliases=None, **kwargs)[source]

Bases: floatParameter

Parameter defined as a read-only function operating on other parameters that returns a float or long double value.

Can access the result of the function through the .quantity attribute, and the value without units through the .value attribute.

On its own this parameter will not be useful, but when inserted into a pint.models.timing_model.Component object it can operate on any parameters within that component or others in the same pint.models.timing_model.TimingModel.

Parameters:

name (str) – The name of the parameter.
func (function) – Returns the desired value
params (iterable) – List or tuple of parameter names. Each can optionally also be a tuple including the attribute to access (default is quantity)
units (str or astropy.units.Quantity) – Parameter default unit. Parameter .value and .uncertainty_value attribute will associate with the default units. If unit is dimensionless, use “’’” as its unit.
description (str, optional) – A short description of what this parameter means.
inpar (bool, optional) – Whether to include in par-file printouts, or to comment out
long_double (bool, optional, default False) – A flag specifying whether value is float or long double.
aliases (list, optional) – An optional list of strings specifying alternate names that can also be accepted for this parameter.

Examples

>>> import pint.models.parameter
>>> p = pint.models.parameter.funcParameter(
        name="AGE",
        description="Spindown age",
        params=("F0", "F1"),
        func=lambda f0, f1: -f0 / 2 / f1,
        units="yr",
    )
>>> m.components["Spindown"].add_param(p)
>>> print(m.AGE)

>>> import pint.models.parameter
>>> import pint.derived_quantities
>>> p2 = pint.models.parameter.funcParameter(
        name="PSREDOT",
        description="Spindown luminosity",
        params=("F0", "F1"),
        func=pint.derived_quantities.pulsar_edot,
        units="erg/s",
    )
>>> m.components["Spindown"].add_param(p2)
>>> print(m.PSREDOT)

Notes

Defining functions through lambda functions may result in unpickleable models

Future versions may include derivative functions to calculate uncertainties.

Methods

`add_alias`(alias)	Add a name to the list of aliases for this parameter.
`as_latex`()
`as_parfile_line`([format])	Return a parfile line giving the current state of the parameter.
`as_ufloat`([units])	Return the parameter as a `uncertainties.ufloat`
`from_parfile_line`(line)	Ignore reading from par file
`from_ufloat`(value[, units])	Set the parameter from the value of a `uncertainties.ufloat`
`help_line`()	Return a help line containing parameter name, description and units.
`name_matches`(name)	Whether or not the parameter name matches the provided name
`prior_pdf`([value, logpdf])	Return the prior probability density.
`set`(value)	Deprecated - just assign to .value.
`str_quantity`(quan)	Quantity as a string (for floating-point values).
`value_as_latex`()

Attributes

`long_double`	Whether the parameter has long double precision.
`params`	Return a list of tuples of parameter names and attributes
`prior`	prior distribution for this parameter.
`quantity`	The result of the function
`repeatable`
`uncertainty`	Parameter uncertainty value with units.
`uncertainty_value`	Return a pure value from .uncertainty.
`unit_scale`	If True, the parameter can automatically scale some values upon assignment.
`units`	Units associated with this parameter.
`value`	The result of the function without units.

add_alias(alias): Add a name to the list of aliases for this parameter.

as_ufloat(units=None)

Return the parameter as a uncertainties.ufloat

Will cast to the specified units, or the default If the uncertainty is not set will be returned as 0

Parameters:: units (astropy.units.core.Unit, optional) – Units to cast the value
Return type:: uncertainties.ufloat

Notes

Currently ufloat does not support double precision values, so some precision may be lost.

from_ufloat(value, units=None)

Set the parameter from the value of a uncertainties.ufloat

Will cast to the specified units, or the default If the uncertainty is 0 it will be set to None

Parameters:

value (uncertainties.ufloat) –
units (astropy.units.core.Unit, optional) – Units to cast the value

help_line(): Return a help line containing parameter name, description and units.

name_matches(name): Whether or not the parameter name matches the provided name

property prior

prior distribution for this parameter.

This should be a Prior object describing the prior distribution of the quantity, for use in Bayesian fitting.

prior_pdf(value=None, logpdf=False)

Return the prior probability density.

Evaluated at the current value of the parameter, or at a proposed value.

Parameters:

value (array-like or float, optional) – Where to evaluate the priors; should be a unitless number. If not provided the prior is evaluated at self.value.
logpdf (bool) – If True, return the logarithm of the PDF instead of the PDF; this can help with densities too small to represent in floating-point.

set(value): Deprecated - just assign to .value.

str_quantity(quan): Quantity as a string (for floating-point values).

property uncertainty_value

Return a pure value from .uncertainty.

This will be interpreted as having units self.units.

property units

Units associated with this parameter.

Should be a astropy.units.Unit object, or None if never set.

property long_double: Whether the parameter has long double precision.

property unit_scale: If True, the parameter can automatically scale some values upon assignment.

property uncertainty: Parameter uncertainty value with units.

property quantity: The result of the function

property value: The result of the function without units.

property params: Return a list of tuples of parameter names and attributes

from_parfile_line(line)[source]

Ignore reading from par file

For funcParameter , it is for information only so is ignored on reading

as_parfile_line(format='pint')[source]

Return a parfile line giving the current state of the parameter.

Parameters:: format (str, optional) – Parfile output format. PINT outputs in ‘tempo’, ‘tempo2’ and ‘pint’ formats. The default format is pint.
Return type:: str

Notes

Format differences between tempo, tempo2, and pint at [1]