https://github.com/supersimple33/tsp_wsp_hardness

Using WSPs to solve the TSP
https://github.com/supersimple33/tsp_wsp_hardness

python travelling-salesman-problem well-seperated-pairs

Last synced: about 1 month ago
JSON representation

Using WSPs to solve the TSP

• Host: GitHub
• URL: https://github.com/supersimple33/tsp_wsp_hardness
• Owner: supersimple33
• Created: 2023-04-28T05:41:57.000Z (over 1 year ago)
• Default Branch: main
• Last Pushed: 2024-05-30T23:12:42.000Z (6 months ago)
• Last Synced: 2024-05-31T01:16:35.708Z (6 months ago)
• Topics: python, travelling-salesman-problem, well-seperated-pairs
• Language: Jupyter Notebook
• Homepage:
• Size: 31.6 MB
• Stars: 1
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: readme.md

Awesome Lists containing this project

README

        
# Well Separated Pairs and the Traveling Salesman Problem

### Instructions for Running

>python tsp-nnp.py \   , -d, -bf}> \

>python tsp-nnp.py att48.tsp 5 \

>python tsp-sps.py att48.tsp 1 -pr -b4 \

>python tsp-sps.py vlsi/xqf131.tsp 2 -pkpr \

>python tsp-spop.py vlsi/xqf131.tsp 1 -pkpmr       <--- Recommended \

>python tsp-treesp.py vlsi/xqf131.tsp 1 -pkpmr     <--- Recommended \

>python wsp-quadtree-test.py att48.tsp 1 -pr -b4 \

>python wsp-quadtree-test.py att48.tsp 1 -pmr

```

quadtrees:

-p: point quadtree 

-pr: point region quadtree

-pmr: PMR quadtree 

-pkpr: PK PR quadtree 

-pkpmr: PK PMR quadtree

flags:

-b: bucket size (for pr and pmr quadtrees), defaults to -b1, ex: -b4

-s: block shrinking on

-d: debug info quadtree and WSP

-bf: brute force, turns off WSP for testing complexity without optimization

```

### Directory Structure

`tsp-bfp.py` - Brute Force + WSP Pruning algorithm - Exponential \

`tsp-nnp.py` - Nearest Neighbor - Polynomial \

`tsp-sps.py` - WSP Subproblem Sorting + Nearest Neighbor (try spop and treesp as they are better) - Polynomial \

`tsp-spop.py` - WSP Subproblem Connection + Brute Force + Nearest Neighbor - Polynomial \

`tsp-treesp.py` - Quadtree Subproblem Connection + Brute Force + Nearest Neighbor - Polynomial \

`tsp-hardness.py` - Compute heuristic to help understand 'hardness' of groups of subproblems to help identify optimal approximation algorithm \

`tsp-wgmst.py` - WSP Graph MST, builds MST out of graph found from WSP min projections - Motivation: https://www.cs.umd.edu/~mount/Indep/Chaojun_Li/final-rept.pdf \

`tsp-msto.py` - WSP Graph MST only - incomplete optimization approach \

`wsp-quadtree-test.py` - Runs WSP alg only to test on different datastructures/quadtrees \

Point set data is located inside the `/data` directory \

WSP implementation inside `/wsp` directory

### Point data

We use the .TSP format (with NODE_COORD_SECTION only) common among TSP datasets.