With psi4 v1.3.2, the following input:
molecule H {
0 2
H 0 0 0
}
memory 32 GB
set reference uks
set basis def2-qzvpp
E = energy('dsd-blyp')
gives the following error:
$ psi4 -i H.input -o H.bug.output
Traceback (most recent call last):
File "/cvmfs/soft.computecanada.ca/easybuild/software/2020/avx2/MPI/intel2020/openmpi4/psi4/1.3.2/bin/psi4", line 287, in <module>
exec(content)
File "<string>", line 27, in <module>
File "/cvmfs/soft.computecanada.ca/easybuild/software/2020/avx2/MPI/intel2020/openmpi4/psi4/1.3.2/lib//python3.8/site-packages/psi4/driver/driver.py", line 556, in energy
wfn = procedures['energy'][lowername](lowername, molecule=molecule, **kwargs)
File "/cvmfs/soft.computecanada.ca/easybuild/software/2020/avx2/MPI/intel2020/openmpi4/psi4/1.3.2/lib//python3.8/site-packages/psi4/driver/procrouting/proc.py", line 2018, in run_scf
dfmp2_wfn.compute_energy()
RuntimeError:
Fatal Error: There are no occupied orbitals with beta spin.
Error occurred in file: /tmp/ebuser/avx2/PSI4/1.3.2/iomkl-2020a/psi4-1.3.2/psi4/src/psi4/dfmp2/mp2.cc on line: 196
The most recent 5 function calls were:
Printing out the relevant lines from the Psithon --> Python processed input file:
core.IO.set_default_namespace("H")
core.set_memory_bytes(32000000000)
core.set_global_option("REFERENCE", "uks")
core.set_global_option("BASIS", "def2-qzvpp")
--> E = energy('dsd-blyp')
Is there a workaround?
EDIT: for posterity, I’ve edited the input and corresponding output to be a minimum example to reproduce this bug.