Parameter (`pmrf.Parameter`)

class pmrf.parameters.Parameter(value, distribution=None, fixed=False, scale=1.0, name=None)[source]

Bases: Module

A container for a parameter, usually used within a Model.

This class serves as the fundamental building block for defining the tunable or fixed parameters within a paramrf Model and for fitting. It is designed to be a flexible container that behaves like a standard numerical type (e.g., a numpy.ndarray) while holding additional metadata for model fitting and analysis.

Usage

Use in mathematical operations just like a JAX/numpy array.
Parameter objects are JAX PyTrees, compatible with JAX transformations (jit, grad).
Mark as fixed (honoured by fitting and sampling routines).
Associate distributions, specified as numpyro distributions (uniform, normal, etc.).

value

The underlying unscaled value. Automatically converted to a float64 array.

Type:: jnp.ndarray

distribution

The prior distribution associated with this parameter.

Type:: numpyro.distributions.Distribution or None

fixed

If True, the parameter is treated as a constant during optimization/sampling.

Type:: bool

scale

A scaling factor. The effective value used in calculations is value * scale.

Type:: float

name

An optional name for the parameter (marked as static).

Type:: str or None

Examples

import pmrf as prf
import jax.numpy as jnp

# A simple, single-valued parameter, initialized with a float
p1 = prf.Parameter(value=1.0e-12, name='C1')

# This parameter can be used in calculations directly (scaling is done during casting)
impedance = 1 / (2j * jnp.pi * 1e9 * p1)
print(f"Impedance: {impedance}")

# A parameter that is fixed and will not be optimized during a fit
p2 = prf.Parameter(value=50.0, fixed=True, name='Z0')

# A parameter with a uniform distribution
# Factory functions are a convenient way to create these
p3 = prf.Uniform(min=0.9e-9, max=1.1e-9, name='L1')

# The parameter's value is initialized to the mean of the distribution
print(f"Initial value of L1: {p3.value}")

value: Array

distribution: Distribution | None = None

fixed: bool = False

scale: float = 1.0

name: str | None = None

property ndim: int

The number of free dimensions for this parameter.

Returns 0 if the parameter is fixed.

property min: array

The unscaled minimum value of the parameter’s distribution.

This is determined by the MIN_PERCENTILE quantile of the distribution.

Returns:: The minimum value, or -inf if no distribution is set.
Return type:: jnp.array

property max: array

The unscaled maximum value of the parameter’s distribution.

This is determined by the MAX_PERCENTILE quantile of the distribution.

Returns:: The maximum value, or inf if no distribution is set.
Return type:: jnp.array

with_value(value)[source]

Return a copy of the parameter with a new unscaled value.

Parameters:: value (jnp.array) – The new unscaled value to set.
Returns:: A copy of this object with value replaced.
Return type:: Parameter

with_distribution(dist)[source]

Return a copy of the parameter with a new distribution.

Parameters:: dist (numpyro.distributions.Distribution) – The distribution to associate with this parameter.
Returns:: A copy of this object with distribution replaced.
Return type:: Parameter
Raises:: Exception – If dist is not a numpyro Distribution.

flattened(separator='_')[source]

Flatten self into a list of scalar Parameters.

If the internal parameter is scalar, the list will contain self. Otherwise, the parameter is split (de-vectorized) if possible.

Parameters:: separator (str, optional, default='_') – Separator used for naming split parameters (e.g., name_0, name_1).
Returns:: The list of individual parameters.
Return type:: list[Parameter]
Raises:: ValueError – If any internal distributions cannot be de-vectorized.

interpolated(x_old, x_new)[source]

Return a new parameter interpolated to a new domain.

Interpolates both the value and the distribution parameters.

Parameters:

x_old (array_like) – The original domain coordinates.
x_new (array_like) – The new domain coordinates.

Returns:

The interpolated parameter.

Return type:

Parameter

as_fixed()[source]

Return a copy of self with fixed=True.

Returns:: The new, fixed parameter.
Return type:: Parameter

as_free()[source]

Return a copy of self with fixed=False.

Returns:: The new, free parameter.
Return type:: Parameter

copy()[source]

Return a shallow copy.

Returns:: A copy created via dataclasses.replace.
Return type:: Parameter

to_json()[source]

Serialize the parameter to a JSON string.

Returns:: A JSON-formatted string containing value, distribution, fixed, scale, and name.
Return type:: str

classmethod from_json(s)[source]

Deserialize a parameter from a JSON string.

Parameters:: s (str) – The JSON string produced by to_json().
Returns:: A reconstructed Parameter instance.
Return type:: Parameter

__init__(value, distribution=None, fixed=False, scale=1.0, name=None)

Parameters:

value (Any)
distribution (Distribution | None)
fixed (bool)
scale (float)
name (str | None)

Return type:

None

Parameters:

value (Array)
distribution (Distribution | None)
fixed (bool)
scale (float)
name (str | None)

class pmrf.parameters.ParameterGroup(param_names, dist=None)[source]

Bases: object

A metadata class that groups a set of named flat parameters and defines any relationships between them.

Parameters:

param_names (list[str] | dict[str, Parameter])
dist (Distribution | None)

parameter_names

The names of the parameters included in this group.

Type:: list[str]

distribution

An optional joint distribution over the flattened parameters.

Type:: dist.Distribution or None

__init__(param_names, dist=None)[source]

Construct a ParameterGroup.

Parameters:

param_names (list[str] | dict[str, Parameter]) – The names of the flattened parameters (or a mapping to parameters).
dist (numpyro.distributions.Distribution, optional) – An optional joint distribution over the flattened parameters.

parameter_names: list[str]

distribution: Distribution | None = None

property num_flat_params

Number of flattened parameters in the group.

Returns:: The count of names in parameter_names.
Return type:: int

property min: array

The unscaled minimum value of the parameter group’s distribution.

Determined by the MIN_PERCENTILE quantile.

Returns:: The minimum value, or -inf if no distribution is set.
Return type:: jnp.array

property max: array

The unscaled maximum value of the parameter group’s distribution.

Determined by the MAX_PERCENTILE quantile.

Returns:: The maximum value, or inf if no distribution is set.
Return type:: jnp.array

Parameter (pmrf.Parameter)

Usage

Parameter (`pmrf.Parameter`)