Original Reference
https://github.com/buqeye/eigenvector-continuation
which is the code for arXiv:2007.03635,
To use the code in Windows, some modification is necessary.
Environment: Mingw64, gfortran, python, no lapack
(1) change the code of fortran file to use internal subroutine instead of lapack library.
this can be done simple commenting.
(2) compile all fortran source code with f2py
python -m numpy.f2py -c -m rmatrix_f2py rmatrix_f2py.f
python -m numpy.f2py -c -m rmatrix_f2py_complex_potential rmatrix_f2py_complex_potential.f
python -m numpy.f2py -c -m coulomb_Barnett coulomb_Barnett.f
(3) copy created dll files(in each sub-directories) to the source directory.
(4) Now we can use the fortran subroutines in python.
(5) The jupyter notebook works now.
댓글 없음:
댓글 쓰기