Exploring 3D Data with ParaView

ParaView enables real time visualization and exploration of three dimensional data sets. The ParaView software is open-source and multi-platform and contains many tools for both qualitative and quantitative data analysis. In this workshop we will explore an example data set and learn how to use a variety of ParaView's built-in tools. Prior to the workshop, please install ParaView on your computer from their website: https://www.paraview.org/. Bring your computer to the workshop so that we can explore ParaView's features together in real time. No prior knowledge or coding experience is required.

The files in the data/ directory are from Pascal Paschos, formerly with NUIT RCS. These data are from a cosmological simulation and contain 3D volumes of log_10(Temperature) and log_10(Density), respectively.

Walkthrough

Note, you can use ParaView interactively, or through a built-in Python Shell:

• View / Python Shell
• Tools / Start Trace (Select "Show Incremental Trace" to see in realtime). This can be very useful if you want to know the Python commands associated with the point + click events in the GUI so that you can exactly reproduce your analysis later.

We will use ParaView interactively:

1. File / Open the Density.vtk and Temperature.vtk files.
2. Turn the "eye" on to view Density.
3. Change the Representation to Volume, and check that the Coloring is set to Density (may already be set this way by default).
4. Do similarly for Temperature.
5. Change the Color Map for Density and/or Temperature.
   • Coloring / Edit -> "Select a color map from default presets" dropdown menu. (You can also access this from the folder+heart icon on the right side of window.)
   • I kept the blue-red for Temperature but tweaked the curve a bit.
   • I changed the Density to green and inverted (white circle with black triangle).
   • Click Apply when you're happy with the color choices.
6. Add a Slice to the Density (from the top bar).
   • click on Density.vtk and then click on slice icon above it.
   • Grabbing the plane (inside the red circle) moves the origin; grabbing the arrow tilts the plane (i.e., moves the normal).
   • When satisfied, click Apply.
7. Hide the Density and Temperature, to view only the Slice.
8. Add a Contour to the Slice (from the top bar).
   • In Value Range select "add a range of values" (icon that looks like a plot axis).
   • I chose from -2 to 2 with 21 linear steps.
   • (You may need/want to remove the initial value that was created by default, using the "-" button.)
   • Click Apply.
9. Add a Filters / Programmable / Calculator to the Temperature.
   • The original data in this file is log_10(Temperature).
   • I entered 10^Temperature, and used "Linear" as the "Result Array Name". (Click Apply.)
   • To view this new Linear data, change the Representation to Volume, and Coloring to Linear.
10. Add a Clip to the Temperature - Calculator (from the top bar).
    • Copy the same positions from the Density Slice to the Temperature Clip.
    • You may want to move it down just a touch so that the contours are more visible.
    • Change the Representation to Surface (if not already set by default).
    • Change Coloring to Linear and then back to (log_10) Temperature to compare to the original logarithmic data. Choose which Coloring you like best.
    • Click Apply.
11. Select both Density and Temperature (Ctrl + click Density + click Temperature) and add Filters / Programmable / Programmable Filter by .
    • Add the following Script. (You may need to swap inputs order depending on how you loaded the data and/or the order that you clicked in ParaView's browser):

D = inputs[0].PointData['Density']
T = inputs[1].PointData['Temperature']
output.PointData.append(D*T, 'multiply')
output.PointData.append(D/T, 'divide')
output.PointData.append(D+T, 'add')
output.PointData.append(cos(T), 'cosT')
output.PointData.append(10**T, 'linearT')
output.PointData.append(10**D, 'linearD')

12. Modify the view of this Filter.
    • Change the Representation to Volume or Surface.
    • Change the Coloring to one of these new arrays.
    • Apply a Threshold (or other) to this (from the top bar).
13. File / Export Scene
    • I chose an eps format (this would need a lot of work to make it publication ready).
    • You can also export to x3dom, which is a format allowed by certain journals for interactive online figures.
14. With the time remaining, explore these data by adding other Filters, modifying colors, changing views, etc.

A final note on getting data into ParaView

ParaView can read and load a large number of different file types. But I often find that my particular data/file type is either not one that ParaView recognizes or there is some error when trying to load it into ParaView. In my experience, ParaView is most happy with .vtk files. Luckily Python has a few packages that can convert appropriate data into .vtk files. Here is ParaView's recommended method to convert data to .vtk in Python. In the past, I have also successfully used PyEVTK for this task.