Browse files

formatting for README 3

  • Loading branch information...
1 parent 1dbdd7a commit bdd0c7a600c39ef1a888d989d8ad57774f880387 @syhw committed Apr 11, 2012
Showing with 9 additions and 3 deletions.
  1. +9 −3 README.markdown
@@ -30,11 +30,17 @@ See *model.pdf* or [CIG'11 publication](
You will always need one runtime library: probt (probt22 in Makefile, any probt >= 2.2 should do).
depending on how it was compiled, the main of model.cpp (*model* (Ben Weber openings), *mymodel* (my openings, c.f. clustering/), or *tt* (tech trees only) in the Makefile) will learn and exit from file (match-up) XvY.txt if given:
- ./model XvY.txt
+ ./model XvY.txt
or will learn on XvY and test on WvZ if given:
- ./model XvY.txt WvZ.txt
+ ./model XvY.txt WvZ.txt
- ./model lZvT.txt tZvT.txt
+ ./model lZvT.txt tZvT.txt
The prediction will always be against X or W, for instance in ZvT, the predictions will be as a Terran player against a Zerg one (the learning is also of Zerg build/tech trees and openings).
1. To just compile and run the Opening prediction model with Ben Weber's openings labels:

0 comments on commit bdd0c7a

Please sign in to comment.