If you have previously defined the three project components – a Dakota study, a simulation model, and an analysis driver – then we are finally ready to run Dakota using the GUI! To run a Dakota study, first click on your study either in the Dakota Studies table in your project settings editor view, or in the project navigator view. You will be taken to the settings editor view for a single Dakota study:
Click on Run This Dakota Study. You should see Dakota running in the console pane:
You may see the following dialog after trying to run Dakota:
Currently, the GUI relies on Dakota’s tabular output file in order to visually display run results. If you choose “Yes, add tabular output to my study”, the GUI will automatically add the necessary keywords to generate tabular output. If you choose “No, run my study anyway,” Dakota will still run, but you may not have Dakota's run result data extracted into the GUI (Dakota's other generated output files will still be created and collected).
After the Dakota run has completed, you can see the result stored in the Run Results table on the settings editor view for your Dakota study:
Click on the run in the project navigator view on the left to view the results:
From here, you can rename the run to something more specific, add comments about your run, or plot the data.
Also, note the "Generated Dakota Run Files" table. These are all of the new files that Dakota generated after running. You have the option to view any of these files, or even "re-link" to completely different arbitrary files on your system that you want to associate with a given run of Dakota for the purpose of GUI analysis (for instance, plotting).