ModelFitting#

class jdaviz.configs.default.plugins.model_fitting.model_fitting.ModelFitting(*args, **kwargs)[source]#

Bases: PluginTemplateMixin, DatasetSelectMixin, SpectralSubsetSelectMixin, DatasetSpectralSubsetValidMixin, NonFiniteUncertaintyMismatchMixin, AddResultsMixin, TableMixin

See the Model Fitting Plugin Documentation for more details.

Only the following attributes and methods are available through the public plugin API:

show()
open_in_tray()
close_in_tray()
cube_fit Only exposed for Cubeviz. Whether to fit the model to the cube instead of to the collapsed spectrum.
dataset (DatasetSelect): Dataset to fit the model.
spectral_subset (SubsetSelect)
model_component (SelectPluginComponent)
poly_order
model_component_label (AutoTextField)
create_model_component()
remove_model_component()
model_components()
valid_model_components()
get_model_component()
set_model_component()
reestimate_model_parameters()
equation (AutoTextField)
equation_components()
add_results (AddResults)
residuals_calculate (bool) Whether to calculate and expose the residuals (model minus data).
residuals (AutoTextField) Label of the residuals to apply when calling calculate_fit() if residuals_calculate is True.
calculate_fit()
fitted_models()
get_models()
get_model_parameters()
get_fitter_parameter()
set_fitter_parameter()

Public constructor

Attributes Summary

`comp_label`	A trait for unicode strings.
`comp_label_auto`	A boolean (True, False) trait.
`comp_label_default`	A trait for unicode strings.
`comp_label_invalid_msg`	A trait for unicode strings.
`component_models`	An instance of a Python list.
`cube_fit`	A boolean (True, False) trait.
`dialog`	A boolean (True, False) trait.
`display_order`	A boolean (True, False) trait.
`eq_error`	A boolean (True, False) trait.
`equation_components`	List of the labels of model components in the current equation
`fitted_models`	Dictionary of all previously fitted models.
`fitter_component`
`fitter_error`	A trait which allows any value.
`fitter_items`	An instance of a Python list.
`fitter_parameters`	An instance of a Python dict.
`fitter_selected`	A trait for unicode strings.
`form_valid_model_component`	A boolean (True, False) trait.
`model_comp_items`	An instance of a Python list.
`model_comp_selected`	A trait for unicode strings.
`model_components`	List of the labels of existing model components
`model_equation`	A trait for unicode strings.
`model_equation_auto`	A boolean (True, False) trait.
`model_equation_default`	A trait for unicode strings.
`model_equation_invalid_msg`	A trait for unicode strings.
`parameter_hints`	An instance of a Python dict.
`poly_order`	Mixin to handle empty integer field.
`poly_order_invalid_msg`	A trait for unicode strings.
`residuals_calculate`	A boolean (True, False) trait.
`residuals_label`	A trait for unicode strings.
`residuals_label_auto`	A boolean (True, False) trait.
`residuals_label_default`	A trait for unicode strings.
`residuals_label_invalid_msg`	A trait for unicode strings.
`template_file`
`user_api`
`valid_model_components`	List of the labels of existing valid (due to display units) model components

Methods Summary

`calculate_fit`([add_data])	Calculate the fit.
`create_model_component`([model_component, ...])	Add a new model component to the list of available model components
`get_fitter_parameter`(key)	Get the value of a fitter parameter.
`get_model_component`(model_component_label[, ...])	Get a (read-only) dictionary representation of an existing model component.
`get_model_parameters`([model_label, x, y])	Convert each parameter of model inside models into a coordinate that maps the model name and parameter name to a `astropy.units.Quantity` object.
`get_models`([models, model_label, x, y])	Loop through all models and output models of the label model_label.
`reestimate_model_parameters`([...])	Re-estimate all free parameters in a given model component given the currently selected data and subset selections.
`remove_model_component`(model_component_label)	Remove an existing model component.
`set_fitter_parameter`(key, value)	Set the value of a fitter parameter.
`set_model_component`(model_component_label, ...)	Set the value or fixed attribute of a parameter in an existing model component.
`vue_add_model`(event)
`vue_apply`(event)
`vue_reestimate_model_parameters`([...])
`vue_remove_model`(event)

Attributes Documentation

comp_label#: A trait for unicode strings.

comp_label_auto#: A boolean (True, False) trait.

comp_label_default#: A trait for unicode strings.

comp_label_invalid_msg#: A trait for unicode strings.

component_models#: An instance of a Python list.

cube_fit#: A boolean (True, False) trait.

dialog#: A boolean (True, False) trait.

display_order#: A boolean (True, False) trait.

eq_error#: A boolean (True, False) trait.

equation_components#: List of the labels of model components in the current equation

fitted_models#: Dictionary of all previously fitted models.

fitter_component#

fitter_error#: A trait which allows any value.

fitter_items#: An instance of a Python list.

fitter_parameters#

An instance of a Python dict.

One or more traits can be passed to the constructor to validate the keys and/or values of the dict. If you need more detailed validation, you may use a custom validator method.

Changed in version 5.0: Added key_trait for validating dict keys.

Changed in version 5.0: Deprecated ambiguous trait, traits args in favor of value_trait, per_key_traits.

fitter_selected#: A trait for unicode strings.

form_valid_model_component#: A boolean (True, False) trait.

model_comp_items#: An instance of a Python list.

model_comp_selected#: A trait for unicode strings.

model_components#: List of the labels of existing model components

model_equation#: A trait for unicode strings.

model_equation_auto#: A boolean (True, False) trait.

model_equation_default#: A trait for unicode strings.

model_equation_invalid_msg#: A trait for unicode strings.

parameter_hints#

An instance of a Python dict.

One or more traits can be passed to the constructor to validate the keys and/or values of the dict. If you need more detailed validation, you may use a custom validator method.

Changed in version 5.0: Added key_trait for validating dict keys.

Changed in version 5.0: Deprecated ambiguous trait, traits args in favor of value_trait, per_key_traits.

poly_order#: Mixin to handle empty integer field.

poly_order_invalid_msg#: A trait for unicode strings.

residuals_calculate#: A boolean (True, False) trait.

residuals_label#: A trait for unicode strings.

residuals_label_auto#: A boolean (True, False) trait.

residuals_label_default#: A trait for unicode strings.

residuals_label_invalid_msg#: A trait for unicode strings.

template_file = ('/home/docs/checkouts/readthedocs.org/user_builds/jdaviz/envs/3978/lib/python3.11/site-packages/jdaviz/configs/default/plugins/model_fitting/model_fitting.py', 'model_fitting.vue')#

user_api#

valid_model_components#: List of the labels of existing valid (due to display units) model components

Methods Documentation

calculate_fit(add_data=True)[source]#

Calculate the fit.

Parameters:

add_databool: Whether to add the resulting spectrum/cube to the app as a data entry according to add_results.

Returns:

fitted model
fitted spectrum/cube
residuals (if residuals_calculate is set to True)

create_model_component(model_component=None, model_component_label=None, poly_order=None)[source]#

Add a new model component to the list of available model components

Parameters:

model_componentstr: Type of model component to add. If not provided, will default according to model_component.
model_component_labelstr: Name of the model component to add. If not provided, will default according to model_component_label (if model_component_label.auto is True and model_component is passed as an argument, then the default label will be recomputed rather than applying the current value).
poly_orderint: Order of the polynomial if model_component is (or defaults to) “Polynomial1D”. Will raise an error if provided and model_component is not “Polynomial1D”. If not provided, will default according to poly_order.

get_fitter_parameter(key)[source]#

Get the value of a fitter parameter.

Parameters:

keystr

The name of the fitter parameter to retrieve. Available parameters depend on the selected fitter. Common parameters include:

maxiter : Maximum number of iterations (available for most fitters)
filter_non_finiteWhether to filter non-finite values
(available for most fitters)
calc_uncertaintiesWhether to calculate uncertainties
(available for most fitters)

Returns:

valueint, float, bool, or None: The current value of the parameter, or None if the parameter does not exist for the selected fitter.

Examples

>>> from jdaviz import Specviz
>>> specviz = Specviz()
>>> plugin = specviz.plugins['Model Fitting']
>>> plugin.fitter.selected = 'LevMarLSQFitter'
>>> max_iterations = plugin.get_fitter_parameter('maxiter')
>>> print(max_iterations)
100

get_model_component(model_component_label, parameter=None)[source]#

Get a (read-only) dictionary representation of an existing model component.

Parameters:

model_component_labelstr: The label given to the existing model component
parameterstr: Optional. The name of a valid parameter in the model component, in which case only the information on that parameter is returned.

get_model_parameters(model_label=None, x=None, y=None)[source]#

Convert each parameter of model inside models into a coordinate that maps the model name and parameter name to a astropy.units.Quantity object.

Parameters:

model_labelstr: Get model parameters for a particular model by inputting its label.
xint: The x coordinate of the model spaxels that will be returned from get_models.
yint: The y coordinate of the model spaxels that will be returned from get_models.

Returns:

dict:: a dictionary of the form {model name: {parameter name: [[astropy.units.Quantity]]}} for 3d models or {model name: {parameter name: astropy.units.Quantity}} where the Quantity object represents the parameter value and unit of one of spaxel models or the 1d models, respectively.

get_models(models=None, model_label=None, x=None, y=None)[source]#

Loop through all models and output models of the label model_label. If x or y is set, return model_labels of those (x, y) coordinates. If x and y are None, print all models regardless of coordinates.

Parameters:

modelsdict: A dict of models, with the key being the label name and the value being an astropy.modeling.CompoundModel object. Defaults to fitted_models if no parameter is provided.
model_labelstr: The name of the model that will be found and returned. If it equals default, every model present will be returned.
xint: The x coordinate of the model spaxels that will be returned.
yint: The y coordinate of the model spaxels that will be returned.

Returns:

selected_modelsdict: Dictionary of the selected models.

reestimate_model_parameters(model_component_label=None)[source]#

Re-estimate all free parameters in a given model component given the currently selected data and subset selections.

Parameters:

model_component_labelstr or None.: The label given to the existing model component. If None, will iterate over all model components.

remove_model_component(model_component_label)[source]#

Remove an existing model component.

Parameters:

model_component_labelstr: The label given to the existing model component

set_fitter_parameter(key, value)[source]#

Set the value of a fitter parameter.

Parameters:

keystr

The name of the fitter parameter to set. Available parameters depend on the selected fitter. Common parameters include:

maxiter : Maximum number of iterations (available for most fitters)
filter_non_finiteWhether to filter non-finite values
(available for most fitters)
calc_uncertaintiesWhether to calculate uncertainties
(available for most fitters)

valueint, float, or bool

The new value for the parameter. The type should match the parameter’s expected type.

Examples

>>> from jdaviz import Specviz
>>> specviz = Specviz()
>>> plugin = specviz.plugins['Model Fitting']
>>> plugin.fitter.selected = 'LevMarLSQFitter'
>>> plugin.set_fitter_parameter('maxiter', 200)
>>> plugin.set_fitter_parameter('filter_non_finite', False)

set_model_component(model_component_label, parameter, value=None, fixed=None)[source]#

Set the value or fixed attribute of a parameter in an existing model component.

Parameters:

model_component_labelstr: The label given to the existing model component
parameterstr: The name of a valid parameter in the model component.
valuefloat: Optional. The new initial value of the parameter. If not provided or None, will remain unchanged.
fixedbool: Optional. The new state of the fixed attribute of the parameter. If not provided or None, will remain unchanged.

Returns:

updated dictionary of the parameter representation

vue_add_model(event)[source]#

vue_apply(event)[source]#

vue_reestimate_model_parameters(model_component_label=None, **kwargs)[source]#

vue_remove_model(event)[source]#