Hello to everyone,
Could you please help me on this? I will be very descriptive for the sake of understanding.
I want to postprocess the following simulacions on Paraview :
- FD1.sim1
- FD1.sim2
- FD1.sim3
- FD2.sim1
- FD2.sim2
…
I have part of the postprocessing methodology as a .py that run on Python shell. It is applicable to any simulation. For now, the last filter is a Calculator and it shows two columns (col1 and col2). I am trying to find the percentile of a column and create isovolumes with those ranges. I know there are two filter that could help: Compute Quartiles and Descriptive Statistics. None of those provides percentiles. Even if it would do, I would like to automatically use those values as input for the isovolume ranges. The automation should be for all the simulations, considering that percentiles will change from one to anotherr.
An alternative to achieve this last goal was to use Python shell in Paraview. It is possible to compute quantiles as the following example:
import numpy as np
x=np.quantile (array, 0.1)
As far as I understand, to me the main issue is the language interpretation between Paraview and Python libraries. Specifically, I can’t access to calculator.col1 via Python shell. That’s what I need to use as the array input in the quantile function. Once I have that data, still need to go back and create the isovolumes to render.
To do this, i am currently exporting Calculator columns and computing percentiles in matlab. Results are manually imported in Paraview afterwards. I believe there is a more optime way to do it.
My questions to you is:
- Do you know any way to compute percentiles by using filters? If yes, how can I use them to automatically generate isovolumes for those ranges?
- Could you helping me finding out how to read columns from calculator to use as input for python and the take the way back and generate isovolumes with that data?
I tried it by reading celldata and pointdata function amongst others that I can’t remember now but not succeed yet.
I wish you could help me on this.
Thanks in advance.