Open pjknowles opened 4 months ago
Further investigation shows that the problem goes away if all of the coordinates in the xyz file have an explicit .
rather than being integers. I guess it is debatable whether or not xyz files that do not conform to this should be supported.
Thank you for the bug report and the already found solution.
I am pretty sure that this would not have been a problem some time ago. Overall numpy and pandas got much more strict about types over time.
I guess it is debatable whether or not xyz files that do not conform to this should be supported.
I always write my own xyz files with explicit floats, but the library should be either permissive or give better error messages.
I think I will add an explicit cast to float
.
PS:
all of the coordinates in the xyz file have an explicit
.
Probably it is already sufficient if one entry per column has the explicit dot, isn't it?
Yes. It's pretty unlikely that you will get an xyz with no '.' (except hand written like mine) but it could happen if written from python str().
Code Sample, a copy-pastable example if possible
Problem description
It crashes while trying to take the square root of a numpy array:
Expected Output
Output of
cc.show_versions()