is a supplementary evolutionary computation and feature selection add-on for
. An example of a Goldenberry program to optimize a cost function using three diferent EDA widgets is shown in figure below. In that program, widgets are the components or processing units needed to perform the optimization task. Rather than depicting
a data ow between these components, connections represent provision and consumption of services encapsulated as objects, which are associated with each input/output interface. For instance, in this program the
widgets require a function to be optimized that is provided by the
widget through three instances of the CostFunction
object, one per EDA, each in charge of keeping statistics of the number of function evaluations and running times. The EDA components are responsible of setting up the parameters
of their corresponding algorithm and of providing a ready-to-use Optimizer that is in turn, consumed by the
. The latter is in fact responsible of orchestrating the execution of these Optimizers, that is, of running each EDA algorithm (whose parameters, including cost function, must have been already defined and provided in their output interfaces
before execution begins) and also of collecting and visualizing the final results. Parameter settings and data presentation are embodied within the the graphical user interface of the widgets, as we shall depict next sections. Intuitively, this programming
paradigm is very much the same as building a hardware apparatus: the user first gather, connect and set-up the required units before switching-on the resulting device.
Goldenberry is hosted publicly under a GPL license. To install the software follow these steps:
1. Download and install latest Orange 2.7 from:
2. Get the latest Goldenberry release from the Downloads tab in
http://goldenberry.codeplex.com (the download link is located to the
left side of the screen).
3. Follow the installation instructions in the release notes just below the download link.
4. Open the Orange application and look for the \Optimization" and \Learner"
toolbar in the canvas (as shown in next figure).