Ctrl+K

Handling Events During Calculations#

Sometimes, it is useful to be able to monitor the progress or success of a fit in real time for long simulations. RAT can send out ‘events’, which contain different data about the calculation. By writing functions that ‘listen’ to these events, you can use this information to build various kinds of updates.

Registering Event Listeners#

When an event is triggered, the event data will be passed to any function that is registered for that specific event type. The event listener function (also called an event handler) should recieve a single argument which will be different depending on the event type. The example below registers a function to listen for the Plot event:

eventManager.register(eventTypes.Plot, 'updatePlot');

import RATapi as RAT
RAT.events.register(RAT.events.EventTypes.Plot, update_plot)

The event listener function is also shown below, it uses one of the plot functions provided by RAT to display the event data. The plot event data contains the current state of the reflectivity and SLD’s, along with a number of other items which is detailed below.

function updatePlot(eventData)

    h = figure(1000);             % Select / open the figure

    subplot(1,2,1); cla           % Reflectivity plot panel
    subplot(1,2,2); cla           % SLD plot panel
    plotRefSLDHelper(eventData);      % Use the standard RAT reflectivity plot
    drawnow limitrate             % Make sure it updates

end

import matplotlib.pyplot as plt

def update_plot(event_data):
    figure = plt.figure(num=1)

    # Use the standard RAT reflectivity plot
    RAT.plotting.plot_ref_sld_helper(event_data, figure)

Note

A utility function already exists to do live plotting, see Live Updating Plots for more information. This section is illustrative for users that want to write more advanced handlers for events.

Event Types#

There are a few event types that are emitted by RAT and each type provides different data as its argument to the listener function. The following events are emitted by RAT:

  1. Message Event

  2. Plot Event

  3. Progress Event

Register a function as an event listener by providing the event type and the listener function to the register function, as shown below.

eventManager.register(eventTypes.Message, handleEvent);   % Message Event
eventManager.register(eventTypes.Plot, handleEvent);      % Plot Event
eventManager.register(eventTypes.Progress, handleEvent);  % Progress Event

import RATapi as RAT

RAT.events.register(RAT.events.EventTypes.Message, handle_event)   # Message Event
RAT.events.register(RAT.events.EventTypes.Plot, handle_event)      # Plot Event
RAT.events.register(RAT.events.EventTypes.Progress, handle_event)  # Progress Event

Message Event#

The message event contains text output from the calculation, which can inform the users about the current step or convergence of the calculation. This event is supported by all algorithms. The event data is a simple string.

Plot Event#

The plot event contains intermediate results from the calculation. As shown in the example above, a good use case for this data is live plotting while the simulation is running. The frequency of the plot events can be controlled from the controls class, see Frequency of Plot Events from more information. This event is supported by the Simplex and DE algorithms. The event data is a structure with the fields described below:

Fields in plot event#

Field

Type

Description

reflectivity

array of double arrays

The calculated reflectivities

shiftedData

array of double arrays

The data corrected with the scalefactor

sldProfiles

array of double arrays

The calculated SLD profiles

resampledLayers

array of double arrays

The resampled layers

subRoughs

array of doubles

The substrate roughness

dataPresent

array of boolean/logical values

flags indicating which contrast contains data

resample

array of boolean/logical values

flags indicating which contrast was resampled

modelType

string

The model type used for the calculation

contrastNames

array of strings

The name for each contrast which can be used to add a plot legend

Progress Event#

The progress event gives the percentage completion for the calculation, and a title text for the event. This event is only supported by the DREAM algorithm. The event data is a structure with the fields described below:

Fields in progress event#

Field

Type

Description

message

string

The title text for the event

percent

float

The percentage of the calculation completed
On this page