The aim of this project was to add concurrency support to Epsilon. The source code can now be found in the main Epsilon repository. This repository currently hosts resources for evaluating performance, for convenience and reproducibility of experiments. For up-to-date runnable JAR files, please see this shared folder. For sample models used in benchmarks, please see this shared folder. Alterantively, if the site is still working, they were originally sourced from here.
Benchmarks can automatically be generated from evaluation_superscript.py (see the cmd/sh files for example invocation). This is provided as a highly convenient and simple way to reproduce benchmarks without having to manually configure command-line arguments each time.