This package extends
Chairmarks by defining a new
macro @bs that shows statistical results of the execution, much like the
@benchmark macro from
BenchmarkTools.
In fact, the code is just an adaptation of the BenchmarkTools code.
Note that this package re-export the @b and @be macros of Chairmarks:
therefore, there is no need to load also Chairmarks.
julia> using PrettyChairmarks
julia> @bs (rand(10)) sort(_) seconds=3
Chairmarks.Benchmark: 101378 samples with 356 evaluations.
Range (min … max): 55.671 ns … 1.567 μs ┊ GC (min … max): 0.00% … 89.48%
Time (median): 73.284 ns ┊ GC (median): 0.00%
Time (mean ± σ): 77.698 ns ± 32.088 ns ┊ GC (mean ± σ): 0.09% ± 2.82%
▅█▇▆▄▁
▁▁▂▅███████▆▄▃▂▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁ ▂
55.7 ns Histogram: frequency by time 189 ns <
Memory estimate: 144.0 bytes, allocs estimate: 1.