I’m trying to speed up an algorithm that computes various graph statistics. The relevant part of the code is: struct erg{G <: AbstractArray{Bool},F <: Function} <: AbstractGraph{Int} m::G fs::Vector{F} trans::G realnodecov::Union{Dict{String,Array{Float64,1}},Nothing} catnodecov…

Automatically fusing together several for-loops

dlakelan September 5, 2020, 2:17pm 6

It should be possible to get speed ups from parallelism. How did you try doing this before?

Topic		Replies	Views
ANN: Parallel `for` loops in FLoops.jl with composable and extensible fold-based API Package Announcements parallel , multithreading , distributed , loops	17	3496	August 25, 2020
Huge performance improvement by separating function? General Usage	15	1207	August 27, 2018
Parallelizing for loop in the computation of a gradient Performance question	19	2632	February 26, 2018
Trying to identify possible optimizations (or errors) in a graph algorithm Graphs lightgraphs	24	1845	January 31, 2019
Poor performance while multithreading (Julia 1.0) Performance multithreading	28	4036	February 11, 2019