julia代码性能调优（一）——profview

一、包的介绍
它可以展示一个火焰图，让你能够看到代码瓶颈在哪里，以便后期展开优化

【This package contains tools for visualizing profiling data collected with Julia's built-in sampling profiler. It can be helpful for getting a big-picture overview of the major bottlenecks in your code, and optionally highlights lines that trigger garbage collection as potential candidates for optimization.

This type of plot is known as a flame graph. The main logic is handled by the FlameGraphs package; this package is just a visualization front-end.
】

二、包的地址
https://github.com/timholy/ProfileView.jl

三、包的安装

using Pkg
Pkg.add("ProfileView")

四、包的用法

function profile_test(n)
    for i = 1:n
        A = randn(100,100,20)
        m = maximum(A)
        Am = mapslices(sum, A; dims=2)
        B = A[:,:,5]
        Bsort = mapslices(sort, B; dims=1)
        b = rand(100)
        C = B.*b
    end
end

using ProfileView
@profview profile_test(1)  # run once to trigger compilation (ignore this one)
@profview profile_test(10)

五、火焰图说明

整体的图

x轴——时间
y轴——调用的嵌套数

各部分对应的源码

用鼠标悬停查看，各部分图对应的源码是那些，以便改进。下图最费时的地方经查询是一个groupby操作。

image.png

六、改进groupby的部分
……
……
待补