fcimc

Python and Fortran implementations of Fleck and Cummings's implicit Monte Carlo (IMC) scheme, published in Journal of Computational Physics (JCP) in 1971.

For each implementation, the results from the published paper (Fleck and Cummings 1971) are reproduced via a set of predefined runs included in the repository.

Dependencies

Python implementation

• Python 3
• Numpy

Fortran implementation

• GFortran

Bundled calculations

• Matplotlib
• Jupyter

Documentation

• Doxygen
• Graphviz
• Doxypypy
• LaTeX
• ghp-import

The local py_filter script must be on the $PATH.

Installation

To create a Conda env with the necessary dependencies:

conda env create -f environment.yml
conda activate fcimc

Add py_filter to the $PATH:

export PATH=".:${PATH}"

Usage

Execution

To build and run everything from scratch, run the top-level script:

./runall clobber  # Removes latest graphs
./runall

To see the aggregated output:

jupyter notebook verification.ipynb

To open the documentation:

open docs/html/index.html

Cleaning

To clean up intermediate build files etc.:

./runall clean

and to clean out everything (output files, executables, etc.) for a clean start:

./runall clobber

Fine control

Each part of the project has its own Makefile which can be invoked directly:

./fortran/src/Makefile
./fortran/calcs/Makefile
./python/calcs/Makefile
./docs/Makefile

to either clean up one part, e.g.:

make -C fortran/src clean
make -C docs clobber

or build/execute, e.g.:

make -C python/calcs -j 4 all

Building the documentation is a special case:

PATH="$PWD/docs:$PATH" make -C docs html

which makes the py_filter command available.

GitHub Pages

The html docs produced by Doxygen can be uploaded to GitHub Pages with:

make -C docs github

(This step is not run by runall.) This uses ghp-import to commit the built documentation to a branch called gh-pages, and push this to GitHub. If the GitHub repository is configured correctly then the gh-pages branch will be rendered as a static site at https://<user>.github.io/fcimc.

Verification

Run the Jupyer notebook to plot the results:

jupyter notebook verification.ipynb

Verification against the original Fleck and Cummings (1971) results (shown on the left) of both a Fortran (middle) and a Python (right) implementation of the IMC scheme described therein.