Dimitrii Lopanov

Convex is done I believe. Now we are doing some other things before we are back to concave polygons.

Hello. I'm glad someone still finds this repository useful :)

Done via #7, but could be refactored due convolutional layer

Trying to do parallelism (not working for currently - WIP): https://github.com/LdDl/ch/tree/optional-parallelism ```shell go.exe test -count=1 -v -timeout 30s -run ^TestParallelShortestPath$ github.com/LdDl/ch ```

> Trying to do parallelism (not working for currently - WIP): https://github.com/LdDl/ch/tree/optional-parallelism > > ```shell > go.exe test -count=1 -v -timeout 30s -run ^TestParallelShortestPath$ github.com/LdDl/ch > ``` Result is right...

Oh well The problem is in _graph.insertShortcuts(batchShortcuts)_ calls. They could not be parallelized since they are inserting incident edges and thus leads us to wrong _dijkstra()_ calls (and this is...

There is no clever trick. This is all about just reducing amount of allocations as you noticed. > reusing the data structures across iterations Yes. It's just used in relaxation...

https://github.com/LdDl/horizon/issues/1 Example of real-world application.

@kkdd Hello and welcome! 1. Routing problem (Dijkstra + Contraction hierarchies): edge relaxation for sure with PQ for eliminating redundant vertices. But I have to mention: it depends on source...

Oh, is that example for road network considering dual roads (e.g. dedicated bus school routes)? *Green - GPS markers, black - just arrowheads of movement direction*  I guess this...