ToDo

Misc

• Patch eko to include beta_1 and revert APFEL
• update regression test
• Make ObservableName class for properly storing "F2light" like objects
• in Runner.__init__: collect default_args in dictionaries (e.g. one for eko stuffs, interpolator-alpha_s-..., one for DIS theory parameters, pto-xiF-xiR-...)
• more metadata in benchmarks' logs

Project management

• update wiki with test description and other useful things for develoment
• restrict Codecov coverage comparison on "feature/" or "release/" branches only (otherways it will always fail simply because benchmarks workflow it's not running on the other branches)

Benchmark Suite

• add options to benchmark (with a local plugin conftest.py) to silence warnings from quad
  • so if we are not interested in the full warning report we can just focus on the new ones and silence the known ones
  • put a hard-coded list where to register warnings to be silenced
• add chance to rerun only failing benchmarks (see docs) (only for local use)

Advertisement

• interesting high dimensional plots: https://plotly.com/python/scientific-charts/ (maybe for db or other kinds of visualization test-related)

Done

• 7/9 points ScaleVariations in the benchmark
• Finish FONLL-A implementation and benchmark with our current APFEL