Download

The MARQOV source code can be obtained from our git repository.

Documentation for the C++ and Python API can be found here: Docs

Compilation

After you have cloned the repository mou might want to compile what you have just obtained.
Since MARQOV utilizes cmake and follows basic cmake procedures,
executing the following in MARQOV’s base folder will suffice to compile all source code that can
be built on your system:

cmake -E make_directory build

cd build

cmake -DCMAKE_BUILD_TYPE=Release ..

cmake --build . --target all --config Release

Then the demos and examples should be present, if available with MPI versions.

Usage

MARQOV is meant to be used as a library.
For a start have a look at the files in the demo subfolder.
There’s mysimpleising and mysimpleheisenberg that show basic usage,
or the numerous example models in the demo subfolder!

Or just hack away to your heart’s content by modifying main.cpp in ./src! There is a Makefile available, such that the compilation is particularly simply. The code can then be executed from any desired working directory and the output as well as log files are placed there (in respective subdirectories). We recommend using the base code folder (one level above the build directory) as the work directory. To compile and execute MARQOV’s main program, use

cd build 
make main
cd ..
./build/src/main

One may avoid switching through directories in the build process using make’s -C flag

make -C build/main
./build/src/main

Supercomputer environments

In order to run MARQOV on larger cluster and supercomputers, we have job scripts available for the Sun Grid Engine (SGE) and SLURM queuing systems. These can be found in the scripts subdirectory.