This is an implementation of HARs: a datastructure (and algorithms) for implementing string diagrams using sparse adjacency matrices.
It accompanies this paper and this talk
For the sparse-matrix-based implementation, see the file cartographer_har/sparse_slice.py
For HAR benchmarks, see the benchmarks folder
For the corresponding catlab benchmarks, see catlab-benchmark.jl
See the ./results folder.
NOTE CAREFULLY: the data in catlab-results.csv
uses time in nanoseconds,
while the data in har-results.csv
has time in seconds.