PyExaFMM CiSE

Setup Environment

Install latest PyExaFMM in a Conda envirionment from source

# Clone
git clone [email protected]:exafmm/pyexafmm.git
cd pyexafmm

# Build
conda build conda.recipe

# Install
conda install --use-local pyexafmm

# Editable mode for live development
python setup.py develop

Appropriate environment variables for PyExaFMM's multithreading implementations can be found in the .env file, and activated with.

source .env

Re-running experiments

All experiments are self contained, and include JSON specifications for re-generating the FMM data structures for a given experiment using PyExaFMM's CLI.

e.g. for a test parametrized with a file named test.json

fmm generate-test-data -c test && fmm compute-operators -c test

Compile manuscript

Requires latex.

cd article && sh compile.sh