qutipf90mc

A "wave-function monte carlo" solver written in Fortran 90/95, with a python interface trough f2py. The program is intended to be used with the QuTiP python package.

Features:

• Usage (almost, see missing features) identical to QuTiP's mcsolve
• Uses sparse (compressed row format) matrices for operators
• Uses zvode to integrate in time
• Time evolution algorithm from QuTiP to find correct times for jumps.
• Automatic parallelization via Python's multiprocessing module.

Missing features:

• Does not accept list as "ntraj" argument.
• Only solves prolbems without explicit time-dependence.

Dependencies:

• QuTiP v.3.1.0 or higher
• A fortran compiler and the BLAS library. BLAS comes with many fortran compilers, such as gfortran. Unoffifical gfortran binaries can be found here https://gcc.gnu.org/wiki/GFortranBinaries

Platforms:

Has been tested on OSX with anaconda python 3.5, gfortran 6.3 and qutip v4.2.0.

Installation:

1. Download code with

git clone https://github.com/arnelg/qutipf90mc.git

2. Enter directory and install

python setup.py install

Or, if you prefer to install locally:

python setup.py build_ext --inplace

Testing and usage:

Test the installation by leaving the directory, starting python and entering

import qutipf90mc

To compare the speed of mcsolve_f90 vs. mcsolve for a decaying system with Hilbert space dimension dim, and ntraj trajectories, run on a single CPU.

qutipf90mc.compare.run(dim,ntraj)

For more info

help(qutipf90mc.mcsolve_f90)

License

You are free to use this software, with or without modification, provided that the conditions listed in the LICENSE.txt file are satisfied.

Contributors

A. L. Grismmo, P. D. Nation, and J. R. Johansson