OpenTURNS Integration in ParaView : Stats and Uncertainty tools

Thanks to EDF funding, new capabilities for data analysis is making its debut in ParaView and VTK based on OpenTURNS. OpenTURNS is an open source initiative for the Treatment of Uncertainties, Risks’N Statistics, and is now an optional dependency of VTK. It takes the form of an optional VTK module that can be requested to be built in ParaView by turning ON the CMake option PARAVIEW_USE_OPENTURNS. Please make sure to have OpenTURNS 1.8 or a more recent version installed on your machine.

OpenTURNS is a scientific C++ and Python library including an internal data model and algorithms dedicated to the treatment of uncertainties. The main goal of this library is to give all the functionalities needed to treat uncertainties in studies to specific applications. Targeted users are engineers who want to introduce the probabilistic dimension in their so far deterministic studies. Most OpenTURNS users use the python API, and we are one of the pioneers to use the C++ API. It is to be noted that OpenTURNS developers, especially Regis LeBrun, helped us with this integration so many thanks to them.

When OpenTURNS module is compild with ParaView, it enables the following new features:

• A new filter “OT Kernel Smoothing” which enables to compute non-parametric curves with different PDF kernel smoothing: Gaussian, Triangular, Epanechnikov. The results can be displayed on top of an histogram:
  

• A new filter “OT Density Map” which enables to compute 2D density map of 2D point clouds, in order to be displayed on top of the point clouds in a Line Chart View:
  

• “Show Active Plot Density Map” and “Show Scatter Plot Density Map” parameters appears in the representation of the PlotMatrixView in order to integrate the “OT Density Map” filter directly into the Plot Matrix View, enabling to show the density map of the first and last decile as well as the median. A few other parameters have been added in order to control color and thickness of the curves:
  

A state file example (.pvsm) can be found here, as some visualizations are not easy to set up.