Python binding

The Python binding is derived directly from the corresponding Fortran binding using the f2py package. So it is not strange that they are almost identical. However, due to the limitations of the f2py package, you have to pay special attentions to the following three APIs:

• cat_init_atomic()
• cat_init_ctqmc()
• cat_init_hfqmc()

In the Fortran binding, the input parameters for the above subroutines are predefined Fortran types. But unfortunately, the f2py does not support them. So we have to customize a f2py-compatible version of them. In the following, we will only show the f2py-compatible version of them. As for the other subroutines, please check section Fortran binding.

> JASMINE component (atomic eigenvalue problem solver)

The definition for the cat_init_atomic() subroutine is as follows:

!!>>> cat_init_atomic: initialize the atomic eigenvalue problem solver
!!>>> python version
  subroutine cat_init_atomic()
     return
  end subroutine cat_init_atomic

NOTE:

Now the user has to setup control parameters for the JASMINE component via the atom.config.in file. With the help of script/u_atomic.py, it is a trivial task.

> CT-HYB quantum impurity solvers

The definition for the cat_init_ctqmc() subroutine is as follows:

!!>>> cat_init_ctqmc: initialize the ctqmc quantum impurity solver
!!>>> python version
  subroutine cat_init_ctqmc(my_id, num_procs)
     integer, intent(in) :: my_id
     integer, intent(in) :: num_procs
     return
  end subroutine cat_init_ctqmc

NOTE:

Now the user has to setup control parameters for the CT-HYB components via the solver.ctqmc.in file. With the help of script/u_ctqmc.py, it is a trivial task.

> HF-QMC quantum impurity solver

The definition for the cat_init_hfqmc() subroutine is as follows:

!!>>> cat_init_hfqmc: initialize the hfqmc quantum impurity solver
!!>>> python version
  subroutine cat_init_hfqmc(my_id, num_procs)
     integer, intent(in) :: my_id
     integer, intent(in) :: num_procs
     return
  end subroutine cat_init_hfqmc

NOTE:

Now the user has to setup control parameters for the HF-QMC components via the solver.hfqmc.in file. With the help of script/u_hfqmc.py, it is a trivial task.