incrementalLHS

Description

This node can be used to generate a series of different plots based on a Dakota incremental LHS study.

Notes

  • LHS stands for “Latin hypercube sampling.”

  • This node is specifically tailored to pull data from Dakota-generated HDF files. It will not work with other types of Dakota output files, nor will it work with arbitrary HDF databases.

  • In addition, only the Dakota sampling method is compatible with this node.

  • This node is solely responsible for writing the plot files to disk. It is not necessary to use file nodes in your workflow to save the plots.

  • After generating and saving the plot file(s), you are free to further modify the plot using the “Chartreuse > Edit plot” context menu option.

Properties

  • methodId: The Dakota method ID used to locate the correct discrete state set variables. If the Dakota study only contained one method block, this field can be left blank.

  • outputFilePrefix: Use this field to apply a common prefix to each generated plot file. This can help to prevent unintentional overwriting of plot data if the workflow is run multiple times.

  • outputPlotFiles: If checked, this node will generate .plot files for each piece of information about the incremental LHS study. These plot files can be subsequently edited after generation.

  • outputPngScreenshots: If checked, this node will generate .png image files for each piece of information about the incremental LHS study.

  • activeVariablesOnly: If checked, then only consider active Dakota variables when creating plots.

  • convergencePlot: If checked, then generate convergence plots. Convergence plots visually demonstrate how the error range of LHS shrinks as the number of refinement samples is incremented.

  • plotPdf: If checked, then generate probability density function histograms for the study.

  • plotCdf: If checked, then generate cumulative probability function histograms for the study.

  • plotScatter: If checked, then generate scatter plots for every unique pairwise combination of Dakota variables and responses.

  • plotCorrelation: If checked, then create correlation coefficient matrix plots. For incrementalLHS, this implies new subfolders containing correlation coefficient matrices, with one subfolder per increment.

  • targetPdfBinCount: Specify a “recommended” number of bins for your PDF plots. Most underlying plotting libraries use a smart binning algorithm based on the data, so this value is just a guideline.

  • targetCdfBinCount: Specify a “recommended” number of bins for your CDF plots. Most underlying plotting libraries use a smart binning algorithm based on the data, so this value is just a guideline.

Input Ports

  • hdfFile: the Dakota-generated HDF input file.

Output Ports

None.