CSC calculation for 2 impurities

[quantumable]

Dear Alex,
following your indication of our last exchange I used the script dmft here{dmft.py.zip} to make the calculation dmft which besides worked by making a direct calculation but if I restore the calculation with the same file .h5 I have an error at this line: S[icrsh].Sigma_iw = ar['DMFT_results']['Iterations']['Sigma_it'+str(iteration_offset-1)+'_'+str(icrsh)]

even doing the csc calculation with the script here{dmft_csc.py.zip} i have the same error. please can you look at the script and tell me what changes to make?

Thank you.

[the-hampel]

Hi,
could you please post the error here and describe your problem? I don’t have time to run your scripts and check for errors. You have to provide more information.

• Alex

[quantumable]

this is the error message:

Starting run with 24 MPI rank(s) at : 2022-03-16 07:57:07.011049
Traceback (most recent call last):
  File "/a/lib/python3.8/runpy.py", line 194, in _run_module_as_main
    return _run_code(code, main_globals, None,
  File "/a/lib/python3.8/runpy.py", line 87, in _run_code
    exec(code, run_globals)
  File "/TRIQS/install/lib/python3.8/site-packages/triqs_dft_tools/converters/plovasp/sc_dmft.py", line 257, in <module>
    main()
  File "/install/lib/python3.8/site-packages/triqs_dft_tools/converters/plovasp/sc_dmft.py", line 254, in main
    run_all(vasp_pid, dmft_mod.dmft_cycle, cfg_file, n_iter, n_iter_dft, vasp_version)
  File "/TRIQS/install/lib/python3.8/site-packages/triqs_dft_tools/converters/plovasp/sc_dmft.py", line 139, in run_all
    corr_energy, dft_dc = dmft_cycle()
  File "/free/csc.py", line 71, in dmft_cycle
    S[icrsh].Sigma_iw = ar['DMFT_results']['Iterations']['Sigma_it'+str(iteration_offset-1)+'_'+str(icrsh)]
IndexError: list index out of range
--------------------------------------------------------------------------
MPI_ABORT was invoked on rank 0 in communicator MPI_COMM_WORLD
with errorcode 1.

NOTE: invoking MPI_ABORT causes Open MPI to kill all MPI processes.
You may or may not see output from other processes, depending on
exactly when Open MPI kills them.
--------------------------------------------------------------------------

[the-hampel]

Hi,
the error is pretty self-explanatory or not? The variable icrsh of your solver list S[icrsh]is out of range. Meaning that you tried to access a impurity problem icrsh for which no solver has been initialized. So you have to make sure that you initialize the solver object before correctly.

[quantumable]

when I do a direct dmft calculation it runs fine but if I restart the calculation using the existing .h5 file it shows me this error! yes I have initialized icrsh with for icrsh in range(SK.n_inequiv_shells):

      for icrsh in range(SK.n_inequiv_shells):
            S[icrsh].Sigma_iw = ar['DMFT_results']['Iterations']['Sigma_it'+str(iteration_offset-1)+'_'+str(icrsh)]

I have attached the two scripts.

[quantumable]

hello sir,
please have you found a idea?

thank you

[the-hampel]

As I said you have to define your solver object earlier if you want to load something into it. Currently you are creating the solver in line 107 of dmft.py:

S.append(Solver(beta=beta, gf_struct=gf_struct))

but you try to load already in line 71 to write into it:

S[icrsh].Sigma_iw = ar['DMFT_results']['Iterations']['Sigma_it'+str(iteration_offset-1)+'_'+str(icrsh)]

This cannot work. You have to create the solver object earlier to access it. Before you write something into it.

• Alex