Skip to content

k-means clustering for poker hands based on earth mover's distance (EMD)

License

Notifications You must be signed in to change notification settings

sammiya/poker-hand-clustering

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

3 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

poker-hand-clustering

MIT License CircleCI

k-means clustering for poker hands based on earth mover's distance (EMD)

About

An implementation of hand abstraction algorithm for two-player nolimit Texas Hold’em poker introduced in Potential-Aware Imperfect-Recall Abstraction with Earth Mover’s Distance in Imperfect-Information Games.

Requirements

  • g++ ( >=7.4 )
  • make
  • cmake (Just for building googletest)

Usage

Please don't forget to run git submodule update --init --recursive after checking out. Then run the clustering program as follows:

make
bin/poker-hand-clustering

In data folder, turn_kmeans.csv and flop_kmeans.csv will be generated. Here we represent cluster by hand indexes implemented in kdub0/hand-isomorphism. Thanks.

CAUTION

This is a rewrite of my old-fashioned code to release, but whether this program works is not yet confirmed enough because of its requirement of computational resource etc. In the future, I will add more tests and run this program with enough computational resource to confirm it.

About

k-means clustering for poker hands based on earth mover's distance (EMD)

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published