Closed javialra97 closed 4 weeks ago
Hi @javialra97 – thanks a lot for reporting this 😄
Is the break
necessary? As with it I think coordinate extraction from optimisation calculations won't grab the last (i.e. optimised) set of coordinates?
Hi @t-young31
True, it is not necessary. I put the break
because I was only associating it with the frequency calculation, but yes, you won't grab the last set of coordinates. Also, I thought that you extracted the geometry from the final Gaussian block.
Describe the bug Sometimes the Gaussian doesn't print the molecule's geometry as 'Input orientation' and only prints the 'Standard orientation'. In these cases, autodE is not able to extract the geometry (L549, autode/wrappers/G09, func coordinates_from) and returns an AssertionError. I have seen these errors mainly during the initial frequency calculation, before the optimization of a TS guess.
To Reproduce Script to launch it:
Error:
File "/home/javialra/anaconda3/envs/autode_original/lib/python3.10/site-packages/autode/wrappers/G09.py", line 675, in hessian_from atoms=self.atoms_from(calc), File "/home/javialra/anaconda3/envs/autode_original/lib/python3.10/site-packages/autode/wrappers/methods.py", line 250, in atoms_from atoms.coordinates = self.coordinates_from(calc) File "/home/javialra/anaconda3/envs/autode_original/lib/python3.10/site-packages/autode/atoms.py", line 641, in coordinates assert value.shape == (len(self), 3) AssertionError
If it is necessary I can provide the full output.
Expected behavior Successful termination
Environment
Additional context To work around, I added the 'Standard orientation' in the if statement (L549, autode/wrappers/G09, func coordinates_from) and then a break at the end.
PS: Amazing work with autodE!!