gnugo-devel
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[gnugo-devel] nonstandard test files


From: Gunnar Farnebäck
Subject: [gnugo-devel] nonstandard test files
Date: Sun, 21 Nov 2004 05:10:57 +0100
User-agent: EMH/1.14.1 SEMI/1.14.3 (Ushinoya) FLIM/1.14.2 (Yagi-Nishiguchi) APEL/10.3 Emacs/21.3 (sparc-sun-solaris2.9) MULE/5.0 (SAKAKI)

We currently have 15 "nonstandard" regression test files in the
distribution, i.e. ones which are not included in the usual regression
runs. It is not clear that we should keep all of them around.

strategy_rot.tst
semeai_rot.tst
score_rot.tst
reading_rot.tst
optics_rot.tst
ld_owl_rot.tst
endgame_rot.tst
connect_rot.tst
connection_rot.tst

Do these fill any purpose anymore? Most of them are empty and none has
been updated for almost two years. I propose removing them. There is
also one file, owl_rot.tst, which is included in the standard
regressions but it's not clear that it's useful either.

semeai2.tst

This is a cut-and-paste collection of tests from standard test files
which are all related to semeai. The main problem with this file is
that the results don't get updated when the original tests do and that
there is a repeated test number that confuses some test scripts.

In any case this can today be simulated by:

regress.pike arb:104 arend:30 buzco:6 dniwog:5 golife:1 golife:2 manyfaces:7 
neurogo:11 neurogo:12 nicklas1:1213,1405,1406,501 nicklas2:1407,904 
strategy:1,37,44 strategy2:66,72,73,80,86,93 
strategy3:109,110,113,124,126,128,129,139,145,146 
strategy4:156,163,168,179,185,199,200,206 trevora:530

and we automatically get up to date results. I propose removing semeai2.tst.

score.tst
score2.tst

I use these from time to time.

newscore.tst

I don't know if anyone actually cares about this file but it's at
least potentially useful.

endgame1.tst

This is only useful together with a removed sgf file. I propose
leaving it in CVS but don't include it in the distribution.

cutstone.tst

These tests don't look very interesting. I propose removing the file.

/Gunnar




reply via email to

[Prev in Thread] Current Thread [Next in Thread]