An experimental CPU/GPU-agnostic Monte Carlo integrator
Switch branches/tags
Nothing to show
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Failed to load latest commit information.
examples
feynman
share
testing
.gitignore
INSTALL.md
README.md
distribute_setup.py
feynman.sublime-project
genint.py
setup-external-posix.sh

README.md

feynman

An experimental Monte Carlo integration framework that allows integrands to be evaluated on either a CPU (using GSL) or a GPU via the same integration API. The original goal was to implement VEGAS GPU integration for Matrix Element Method analyses, but sadly I ran out of time to make that a reality... Plain Monte Carlo integration is implemented on the GPU, and even works for some sample integrands, but VEGAS is not, so Matrix Element integration would likely be too slow to be practical (Matix Element phase space is very spiky).

This code is likely not fit for any use, but might be a useful reference in the future.