PkgBenchmark.jl
PkgBenchmark.jl copied to clipboard
Easy benchmark tracking for packages
sometimes one just want to judge ``` @time main_function_for_user(...) ```
Hi, As explained on [Discourse](https://discourse.julialang.org/t/how-to-compare-commits-with-pkgbenchmark/104468/3), I was unable to run a benchmark because the package was not added as `dev`. As far as I can tell, the documentation does not...
I'm trying to run a benchmark against master vs PR where the PR bumps a dependency version one breaking level up. But it fails to checkout the older version on...
This allows to get updates for GitHub actions automatically - it is basically an equivalent of CompatHelper. I have used this for my own packages, the [Trixi.jl framework](https://github.com/trixi-framework), and the...
I am running `PkgBenchmark.jl` using the [test file](https://github.com/JuliaCI/PkgBenchmark.jl/blob/master/benchmark/benchmarks.jl). Then, ``` commit = benchmarkpkg(repo_name) # current state of repository main = benchmarkpkg(repo_name, "main") judgement = judge(commit, main) commit_stats = bmark_results_to_dataframes(commit) main_stats...
Add an explanation of percentages in brackets in the markdown output of judgments and a link to the `BenchmarkTools.jl` manual page describing `time_tolerance` and `memory_tolerance`.
changing `Base.strwidth` to `Base.textwidth`, which is the current equivalent method
I ran into an error while benchmarking that was pretty hard to track down: ``` ERROR: /home/runner/work/MyProject.jl/MyProject.jl is dirty. Please commit/stash your changes before benchmarking a specific commit ``` But...
Is there a way to benchmark an import of a package, i.e. something similar to ```julia julia>SUITE = BenchmarkGroup() julia>SUITE["Loading SetTools"] = @benchmarkable using MyPackage ERROR: syntax: "import" expression not...
Hey! First of all, thanks for providing this nice package, it makes comparing of several commits really easy :) I've tried to judge benchmarks for several julia versions. Unfortunately, this...