Q2MM / q2mm

Quantum to Molecular Mechanics (Q2MM)
MIT License
24 stars 9 forks source link

Error running a loop for FF optimization on seminario branch #77

Open landreola opened 8 months ago

landreola commented 8 months ago

Description

When running a normal optimization loop while my q2mm version was set to the seminario branch, I ran into an error that did not occur after I switched to the master branch. This error caused the program to crash and not return the opt file.

Traceback

Traceback (most recent call last):
  File "/afs/crc.nd.edu/user/l/landreol/q2mm/q2mm/opt.py", line 36, in wrapper
    return func(*args, **kwargs)
  File "/afs/crc.nd.edu/user/l/landreol/q2mm/q2mm/simplex.py", line 155, in run
    params = select_simp_params_on_derivs(
  File "/afs/crc.nd.edu/user/l/landreol/q2mm/q2mm/simplex.py", line 414, in select_simp_params_on_derivs
    calc_simp_var(params)
  File "/afs/crc.nd.edu/user/l/landreol/q2mm/q2mm/simplex.py", line 401, in calc_simp_var
    param.simp_var = param.d2 / param.d1**2.
ZeroDivisionError: float division by zero

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "/afs/crc.nd.edu/user/l/landreol/q2mm/q2mm/loop.py", line 430, in <module>
    main(sys.argv[1:])
  File "/afs/crc.nd.edu/user/l/landreol/q2mm/q2mm/loop.py", line 424, in main
    loop.run_loop_input(lines)
  File "/afs/crc.nd.edu/user/l/landreol/q2mm/q2mm/loop.py", line 144, in run_loop_input
    self.ff = loop.opt_loop()
  File "/afs/crc.nd.edu/user/l/landreol/q2mm/q2mm/loop.py", line 64, in opt_loop
    self.ff = self.run_loop_input(
  File "/afs/crc.nd.edu/user/l/landreol/q2mm/q2mm/loop.py", line 383, in run_loop_input
    self.ff = simp.run(r_data=self.ref_data)
  File "/afs/crc.nd.edu/user/l/landreol/q2mm/q2mm/opt.py", line 48, in wrapper
    papa_bear.best_ff.export_ff(papa_bear.best_ff.path)
AttributeError: 'list' object has no attribute 'export_ff'

To Reproduce

You should be able to reproduce it by running run.job in the compressed folder, but it did take a few hours to crash out. The output of the run is also in the folder and is named run.o194770. I am guessing it would also happen if you reduced the number of structures used in the optimization, but I did not try that. ir_hydrog_seminario_error.tar.gz