You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Turing's benchmarking is not used often at the moment. I think part of the reason is that the benchmarking process is a bit complicated, even with Nanosoldiers. Would it make sense to try to
trigger benchmarking for PRs using bors bot, with a customised command (see below)
replace Nanosoldier with PkgBenchmark, which is designed specifically for track performance changes of Julia packages.
Ps: some references:
bors tryarguments: these arguments are stored as commit message, and can be extracted by the benchmarking script.
perhaps: bors try --benchmarking?
The text was updated successfully, but these errors were encountered:
Maybe https://github.com/tkf/BenchmarkCI.jl could be useful. The trigger of the Github action could be adjusted to whatever we think would be suitable.
Turing's benchmarking is not used often at the moment. I think part of the reason is that the benchmarking process is a bit complicated, even with Nanosoldiers. Would it make sense to try to
PkgBenchmark
, which is designed specifically for track performance changes of Julia packages.Ps: some references:
bors try
arguments: these arguments are stored as commit message, and can be extracted by the benchmarking script.bors try --benchmarking
?The text was updated successfully, but these errors were encountered: