You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Occasionally, FindShapes gives a NaN axis Euler angle output. This happens when uberbelim[elimcount - 1][0] becomes a NaN value due to uberelim[elimcount - 1][elimcount - 1] being zero.
In my case, the NaN values occur when the calculated moments of a specific grain are Ixx=4.5, Iyy=3.5=Izz=3.5, Ixy=Iyz=Ixz=-1, and the resulting uber matrix for j=1 is {{0, -1, -1}, {-1, -1, -1}{-1, -1, -1}}. Here is a picture of this "failed" grain for what is worth:
I'm not sure what the true root cause of this is, though.
The text was updated successfully, but these errors were encountered:
Thank you for the test base. The FeatureId that has the issue is 219. Debugging through the case we end up in a case with a 0/0 which will give NaN....
in the function void FindShapes::find_axiseulers() if Ixy, Iyz, Ixzall equal -1 then NaN values will be produced for the axis Euler values. this seems to happen with the feature is in the shape from @cartercocke original post above. I'm not sure if this is an edge case or if there is a subtle error in the codes. Further investigation is needed.
This image shows 4 features from the sample data set that all have the same values for Ixy, Iyz, Ixz but 26 and 126 will not produce NaN values where as 219 and 261 will produce NaN values. Attached image shows the details.
There is probably an edge case that is not taken into account.
Occasionally, FindShapes gives a NaN axis Euler angle output. This happens when
uberbelim[elimcount - 1][0]
becomes a NaN value due touberelim[elimcount - 1][elimcount - 1]
being zero.DREAM3D/Source/Plugins/Statistics/StatisticsFilters/FindShapes.cpp
Line 709 in a3f74d7
In my case, the NaN values occur when the calculated moments of a specific grain are Ixx=4.5, Iyy=3.5=Izz=3.5, Ixy=Iyz=Ixz=-1, and the resulting
uber
matrix forj=1
is {{0, -1, -1}, {-1, -1, -1}{-1, -1, -1}}. Here is a picture of this "failed" grain for what is worth:I'm not sure what the true root cause of this is, though.
The text was updated successfully, but these errors were encountered: