Error - RBF Deformation of UNV

[prosimas]

In the latest version there seems to be an issue with RBF deformation of unv files. Loaded a case which was working 2 months ago, but doesn't in the latest update.

The error comes up, while reading the rbf parameter file, telling me that the number of control points is unequal to deformed points. In fact: number is the same. If I comment the typeerror in the params/rbf.py the error doesn't show up, but the geometry isn't deformed either.

What's wrong with latest rbf optimization?

[prosimas]

By the way: great project! 👍

[mtezzele]

Hi. Can you provide here the problematic parameter file?

[prosimas]

Hi Marco,

thank you for your fast reply! The parameter file is generated in matlab/octave by a script I wrote to extract the basis points from arbitrary geometries. This is working perfectly with the older PyGeM-version. Here is the parameter file in question:

`[Radial Basis Functions]
basis function: beckert_wendland_c2_basis
radius: 0.50
power: 2

[Control points]
original control points: -0.024899 0.000000 -0.002240
-0.019187 0.000000 0.016020
-0.016020 0.000000 -0.019187
-0.002240 0.000000 0.024901
0.000878 0.049459 -0.009729
0.002240 0.000000 -0.024901
0.005163 0.048575 0.006696
0.007538 0.048153 -0.024994
0.016020 0.000000 0.019187
0.017873 0.046020 0.014662
0.019187 0.000001 -0.016020
0.021232 0.045423 -0.030163
0.024899 -0.000000 0.002240
0.031567 0.043290 0.009493
0.033942 0.042867 -0.022197
0.038227 0.041983 -0.005772
0.038620 0.085105 -0.012051
0.049950 0.078226 0.009592
0.052190 0.077031 -0.030260
0.063520 0.070152 -0.008617
0.088328 0.106936 -0.009206
0.090548 0.103016 -0.021109
0.093993 0.096617 0.009689
0.095112 0.094825 -0.025190
0.098557 0.088427 0.005608
0.099349 0.087160 -0.019050
0.100777 0.084507 -0.006295
0.150000 0.085048 0.001194
0.150000 0.088585 -0.009726
0.150000 0.090274 0.011415
0.150000 0.098805 -0.014953
0.150000 0.101195 0.014953
0.150000 0.109726 -0.011415
0.150000 0.114952 -0.001194

deformed control points: -0.024899 0.000000 -0.002240
-0.019187 0.000000 0.016020
-0.016020 0.000000 -0.019187
-0.002240 0.000000 0.024901
0.000878 0.049459 -0.009729
0.002240 0.000000 -0.024901
0.005163 0.048575 0.006696
0.007538 0.048153 -0.024994
0.016020 0.000000 0.019187
0.017873 0.046020 0.014662
0.019187 0.000001 -0.016020
0.021232 0.045423 -0.030163
0.024899 -0.000000 0.002240
0.031567 0.043290 0.009493
0.033942 0.042867 -0.022197
0.038227 0.041983 -0.005772
0.046344 0.068084 -0.009641
0.039960 0.062581 0.011510
0.041752 0.061625 -0.024208
0.063520 0.070152 -0.008617
0.088328 0.106936 -0.009206
0.090548 0.103016 -0.021109
0.093993 0.096617 0.009689
0.095112 0.094825 -0.025190
0.098557 0.088427 0.005608
0.099349 0.087160 -0.019050
0.100777 0.084507 -0.006295
0.150000 0.085048 0.001194
0.150000 0.088585 -0.009726
0.150000 0.090274 0.011415
0.150000 0.098805 -0.014953
0.150000 0.101195 0.014953
0.150000 0.109726 -0.011415
0.150000 0.114952 -0.001194
`

[mtezzele]

Thanks, I found the bug in the shape initialization of the control points. I will let you know when the PR #150 is merged.

[prosimas]

Great! 👍

[mtezzele]

@prosimas now it should work. Let us know if you still have problems. Otherwise I will close this issue in the next days.

Thanks again for finding the bug!

[prosimas]

Works perfectly, Marco! Thank you and your team for this great project! Soon my website will go online where I will promote PyGeM👍

[mtezzele]

@prosimas fantastic! Let us know when it's ready so we can promote each other.