JuliaPlots / RecipesPipeline.jl

Utilities for processing recipes
http://juliaplots.org/RecipesPipeline.jl/dev/
MIT License
17 stars 17 forks source link

DefaultsDict - correctness and performance tweaks for large numbers of series #126

Closed BioTurboNick closed 2 years ago

BioTurboNick commented 2 years ago

Test example:

const xxxx = rand(10, 10000) # lots of small series
plot(xxxx);

Noticed that the delete! methods were sometimes returning an underlying dictionary rather than the DefaultDict itself; improved type stability resulted in large reduction in allocation count: 825.900 ms (5210177 allocations: 255.31 MiB) to 814.990 ms (4470177 allocations: 232.73 MiB)

codecov-commenter commented 2 years ago

Codecov Report

Base: 62.00% // Head: 62.03% // Increases project coverage by +0.02% :tada:

Coverage data is based on head (4651d59) compared to base (de1d6f2). Patch coverage: 75.00% of modified lines in pull request are covered.

Additional details and impacted files ```diff @@ Coverage Diff @@ ## master #126 +/- ## ========================================== + Coverage 62.00% 62.03% +0.02% ========================================== Files 10 10 Lines 558 561 +3 ========================================== + Hits 346 348 +2 - Misses 212 213 +1 ``` | [Impacted Files](https://codecov.io/gh/JuliaPlots/RecipesPipeline.jl/pull/126?src=pr&el=tree&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=JuliaPlots) | Coverage Δ | | |---|---|---| | [src/utils.jl](https://codecov.io/gh/JuliaPlots/RecipesPipeline.jl/pull/126/diff?src=pr&el=tree&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=JuliaPlots#diff-c3JjL3V0aWxzLmps) | `56.57% <75.00%> (+0.41%)` | :arrow_up: | Help us with your feedback. Take ten seconds to tell us [how you rate us](https://about.codecov.io/nps?utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=JuliaPlots). Have a feature suggestion? [Share it here.](https://app.codecov.io/gh/feedback/?utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=JuliaPlots)

:umbrella: View full report at Codecov.
:loudspeaker: Do you have feedback about the report comment? Let us know in this issue.