autofit.LogGaussianPrior#

class LogGaussianPrior[source]#

Bases: Prior

A prior for a variable whose logarithm is gaussian distributed. Work in natural log.

The conversion of an input unit value, u, to a physical value, p, via the prior is as follows:

\[p = \mu + (\sigma * sqrt(2) * erfcinv(2.0 * (1.0 - u))\]

For example for prior = LogGaussianPrior(mean=1.0, sigma=2.0), an input prior.value_for(unit=0.5) is equal to 1.0.

[Rich describe how this is done via message]

Parameters:

mean (float) – The natural log of the distribution’s mean.
sigma (float) – The spread of this distribution in natural log space, e.g. sigma=1.0 means P(ln x) has a standard deviation of 1.
lower_limit (float) – A lower limit in real space (not log); physical values below this are rejected.
upper_limit (float) – A upper limit in real space (not log); physical values above this are rejected.

Examples

prior = af.LogGaussianPrior(mean=1.0, sigma=2.0, lower_limit=0.0, upper_limit=2.0)

physical_value = prior.value_for(unit=0.5)

Methods

`assert_within_limits`
`dict`	A dictionary representation of this prior
`for_class_and_attribute_name`
`from_dict`	Returns a prior from a JSON representation.
`gaussian_prior_model_for_arguments`
`has`	Does this instance have an attribute which is of type cls?
`instance_for_arguments`
`log_prior_from_value`
`make_indexes`	rtype: `Tuple`[`ndarray`, `...`]
`name_of_class`	A string name for the class, with the prior suffix removed.
`new`	Returns a copy of this prior with a new id assigned making it distinct
`next_id`
`project`
`random`	A random value sampled from this prior
`replacing_for_path`	Create a new model replacing the value for a given path with a new value
`tree_unflatten`	Create a prior from a flattened PyTree
`unit_value_for`	Compute the unit value between 0 and 1 for the physical value.
`value_for`	Return a physical value for a value between 0 and 1 with the transformation described by this prior.
`with_limits`	Create a new instance of the same prior class with the passed limits.
`with_message`

Attributes

`component_number`
`factor`	A callable PDF used as a factor in factor graphs
`identifier`
`label`
`limits`	rtype: `Tuple`[`float`, `float`]
`lower_unit_limit`	The lower limit for this prior in unit vector space
`ndim`	How many dimensions does this variable have?
`parameter_string`	rtype: `str`
`upper_unit_limit`	The upper limit for this prior in unit vector space
`width`

value_for(unit, ignore_prior_limits=False)[source]#

Return a physical value for a value between 0 and 1 with the transformation described by this prior.

Parameters:: unit (float) – A unit value between 0 and 1.
Return type:: A physical value, mapped from the unit value accoridng to the prior.