8.2.2. The Bayesian Extension and utilities

Summary of the main classes for the Bayesian Optimization extension

BayesianOptimisation(dockarea, dashboard)	PyMoDAQ extension of the DashBoard to perform the optimization of a target signal taken form the detectors as a function of one or more parameters controlled by the actuators.
BayesianModelGeneric(optimisation_controller)	Methods
BayesianModelDefault(optimisation_controller)	Methods

8.2.2.1. The Extension module

class pymodaq.extensions.BayesianOptimisation(dockarea, dashboard)

PyMoDAQ extension of the DashBoard to perform the optimization of a target signal taken form the detectors as a function of one or more parameters controlled by the actuators.

Attributes

modules_manager

useful tool to interact with DAQ_Moves and DAQ_Viewers

Methods

connect_things()	Connect actions and/or other widgets signal to methods
setup_actions()	Method where to create actions to be subclassed.
setup_docks()	to be subclassed to setup the docks layout for instance:
setup_menu()	to be subclassed create menu for actions contained into the self.actions_manager, for instance:
value_changed(param)	to be subclassed for actions to perform when one of the param's value in self.settings is changed

clean_h5_temp
command_runner
enable_controls_opti
format_bounds
get_set_model_params
get_stopping_parameters
go_to_best
ini_live_plot
ini_model
ini_optimisation_runner
ini_temp_file
optimisation_done
optimisation_done_signal
process_output
quit
run_optimisation
set_algorithm
set_model
update_actuators
update_bounds
update_data_plot
update_stopping_criteria
update_utility_function

connect_things()

Connect actions and/or other widgets signal to methods

setup_actions()

Method where to create actions to be subclassed. Mandatory

Examples

>>> self.add_action('Quit', 'close2', "Quit program")
>>> self.add_action('Grab', 'camera', "Grab from camera", checkable=True)
>>> self.add_action('Load', 'Open', "Load target file (.h5, .png, .jpg) or data from camera", checkable=False)
>>> self.add_action('Save', 'SaveAs', "Save current data", checkable=False)

See also

ActionManager.add_action

setup_docks()

to be subclassed to setup the docks layout for instance:

self.docks[‘ADock’] = gutils.Dock(‘ADock name) self.dockarea.addDock(self.docks[‘ADock”]) self.docks[‘AnotherDock’] = gutils.Dock(‘AnotherDock name) self.dockarea.addDock(self.docks[‘AnotherDock”], ‘bottom’, self.docks[‘ADock”])

See also

pyqtgraph.dockarea.Dock

setup_menu()

to be subclassed create menu for actions contained into the self.actions_manager, for instance:

For instance:

file_menu = self.menubar.addMenu(‘File’) self.actions_manager.affect_to(‘load’, file_menu) self.actions_manager.affect_to(‘save’, file_menu)

file_menu.addSeparator() self.actions_manager.affect_to(‘quit’, file_menu)

value_changed(param)

to be subclassed for actions to perform when one of the param’s value in self.settings is changed

For instance: if param.name() == ‘do_something’:

if param.value():

print(‘Do something’) self.settings.child(‘main_settings’, ‘something_done’).setValue(False)

Parameters

param ((Parameter) the parameter whose value just changed) –

property modules_manager: ModulesManager

useful tool to interact with DAQ_Moves and DAQ_Viewers

Will be available if a DashBoard has been set

Return type

ModulesManager

8.2.2.2. The Base Models

class pymodaq.extensions.BayesianModelGeneric(optimisation_controller: BayesianOptimisation)

Methods

convert_input(measurements)	Convert the measurements in the units to be fed to the Optimisation Controller :param measurements: data object exported from the detectors from which the model extract a float value (fitness) to be fed to the algorithm :type measurements: DataToExport
convert_output(outputs[, best_individual])	Convert the output of the Optimisation Controller in units to be fed into the actuators :param outputs: output value from the controller from which the model extract a value of the same units as the actuators :type outputs: list of numpy ndarray :param best_individual: the coordinates of the best individual so far :type best_individual: np.ndarray
ini_model()	To be subclassed
runner_initialized()	To be subclassed
update_plots()	Called when updating the live plots
update_settings(param)	Get a parameter instance whose value has been modified by a user on the UI To be overwritten in child class

check_modules
ini_model_base
optimisation_algorithm
update_detector_names

convert_input(measurements: DataToExport) → float

Convert the measurements in the units to be fed to the Optimisation Controller :param measurements: data object exported from the detectors from which the model extract a float value

(fitness) to be fed to the algorithm

Return type

float

convert_output(outputs: List[ndarray], best_individual=None) → DataToActuators

Convert the output of the Optimisation Controller in units to be fed into the actuators :param outputs: output value from the controller from which the model extract a value of the same units as the actuators :type outputs: list of numpy ndarray :param best_individual: the coordinates of the best individual so far :type best_individual: np.ndarray

Returns

DataToActuatorOpti – attribute, either ‘rel’ for relative or ‘abs’ for absolute.

Return type

derived from DataToExport. Contains value to be fed to the actuators with a a mode

ini_model()

To be subclassed

Initialize whatever is needed by your custom model

runner_initialized()

To be subclassed

Initialize whatever is needed by your custom model after the optimization runner is initialized

update_plots()

Called when updating the live plots

update_settings(param: Parameter)

Get a parameter instance whose value has been modified by a user on the UI To be overwritten in child class

class pymodaq.extensions.BayesianModelDefault(optimisation_controller: BayesianOptimisation)

Methods

convert_input(measurements)	Convert the measurements in the units to be fed to the Optimisation Controller
convert_output(outputs[, best_individual])	Convert the output of the Optimisation Controller in units to be fed into the actuators :param outputs: output value from the controller from which the model extract a value of the same units as the actuators :type outputs: list of numpy ndarray :param best_individual: the coordinates of the best individual so far :type best_individual: np.ndarray
ini_model()	To be subclassed
update_settings(param)	Get a parameter instance whose value has been modified by a user on the UI To be overwritten in child class

optimize_from

convert_input(measurements: DataToExport) → float

Convert the measurements in the units to be fed to the Optimisation Controller

Parameters

measurements (DataToExport) – data object exported from the detectors from which the model extract a float value (fitness) to be fed to the algorithm

Return type

float

convert_output(outputs: List[ndarray], best_individual=None) → DataToActuators

Convert the output of the Optimisation Controller in units to be fed into the actuators :param outputs: output value from the controller from which the model extract a value of the same units as the actuators :type outputs: list of numpy ndarray :param best_individual: the coordinates of the best individual so far :type best_individual: np.ndarray

Returns

• DataToActuators (derived from DataToExport. Contains value to be fed to the actuators)

• with a mode attribute, either ‘rel’ for relative or ‘abs’ for absolute.

ini_model()

To be subclassed

Initialize whatever is needed by your custom model

update_settings(param: Parameter)

Get a parameter instance whose value has been modified by a user on the UI To be overwritten in child class