Code

Rustworkx acceleration#381

Code

Draft

InnovativeInventor wants to merge 26 commits intomainmggg/GerryChain:mainfrom retworkxmggg/GerryChain:retworkxCopy head branch name to clipboard

+156-29

Commits

Commits on Apr 25, 2023

• Add __slots__ and replace PopulatedGraph.degree with direct dictionary lookup

  

  InnovativeInventor

  authored and

  pjrule

  committed
  c1ff2f8

  Copy full SHA for c1ff2f8

• Switch to using more performant/preferred lookup call

  

  InnovativeInventor

  authored and

  pjrule

  committed
  b72238f

  Copy full SHA for b72238f

• Add retworkx converter

  

  InnovativeInventor

  authored and

  pjrule

  committed
  3692209

  Copy full SHA for 3692209

• Use graph.node_indicies in PopulatedGraph

  

  InnovativeInventor

  authored and

  pjrule

  committed
  a5e90af

  Copy full SHA for a5e90af

• Add keep_attributes flag (will be released in retworkx)

  

  InnovativeInventor

  authored and

  pjrule

  committed
  ea2cb12

  Copy full SHA for ea2cb12

• Naive bipartition_tree retworkx impl

  

  InnovativeInventor

  authored and

  pjrule

  committed
  a74c6c2

  Copy full SHA for a74c6c2

• Improve subgraph instantiation times

  

  InnovativeInventor

  authored and

  pjrule

  committed
  4f99cd4

  Copy full SHA for 4f99cd4

• More perf improvements

  

  InnovativeInventor

  authored and

  pjrule

  committed
  d3f0dc4

  Copy full SHA for d3f0dc4

• Stop using recursive_tree_part

  

  InnovativeInventor

  authored and

  pjrule

  committed
  190d2ed

  Copy full SHA for 190d2ed

• Revert "Stop using recursive_tree_part"

  Show description for 87b3905
  

  InnovativeInventor

  authored and

  pjrule

  committed
  87b3905

  Copy full SHA for 87b3905

• Resolve memory leak issue with caching

  

  InnovativeInventor

  authored and

  pjrule

  committed
  d9b3c23

  Copy full SHA for d9b3c23

• Prevent unnecessary subgraphing

  

  InnovativeInventor

  authored and

  pjrule

  committed
  73e9990

  Copy full SHA for 73e9990

• Maybe faster record perf (todo: benchmark carefully)

  

  InnovativeInventor

  authored and

  pjrule

  committed
  3b3f36c

  Copy full SHA for 3b3f36c

• Re-enable lookup cache

  

  InnovativeInventor

  authored and

  pjrule

  committed
  9c2f38c

  Copy full SHA for 9c2f38c

• Use dict.update operator to support older versions of Python

  

  InnovativeInventor

  authored and

  pjrule

  committed
  6484374

  Copy full SHA for 6484374

• Fix misspelled node_indicies [sic]

  

  InnovativeInventor

  authored and

  pjrule

  committed
  eeb317b

  Copy full SHA for eeb317b

• Fix linter issues

  

  InnovativeInventor

  authored and

  pjrule

  committed
  f1310d9

  Copy full SHA for f1310d9

• Add retworkx as setup.py dep

  

  InnovativeInventor

  authored and

  pjrule

  committed
  c9d06ba

  Copy full SHA for c9d06ba

• Reduce overhead of to_series call

  

  InnovativeInventor

  authored and

  pjrule

  committed
  063e077

  Copy full SHA for 063e077

• Switch to __getattr__ call to reduce overhead of attr fetches

  

  InnovativeInventor

  authored and

  pjrule

  committed
  762b8ac

  Copy full SHA for 762b8ac

• Force self.mapping to maintain sorted order invariant for faster serialization

  

  InnovativeInventor

  authored and

  pjrule

  committed
  1ea4df9

  Copy full SHA for 1ea4df9

• Update retworkx call to use renamed function

  

  InnovativeInventor

  authored and

  pjrule

  committed
  bd7b604

  Copy full SHA for bd7b604

• Rename bipartition_graph to bipartition_graph_mst

  

  InnovativeInventor

  authored and

  pjrule

  committed
  bc1356d

  Copy full SHA for bc1356d

• Make rustworkx (née retworkx) an optional dependency

  pjrule

  committed
  48801ed

  Copy full SHA for 48801ed

• Revert changes to Partition class

  pjrule

  committed
  37831ba

  Copy full SHA for 37831ba

Commits on Apr 29, 2023

• WIP: Use gerrychain.rs for acceleration

  pjrule

  committed
  1ce30d0

  Copy full SHA for 1ce30d0