[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[gnugo-devel] regression patch
From: |
Gunnar Farneback |
Subject: |
[gnugo-devel] regression patch |
Date: |
Wed, 17 Oct 2001 22:30:53 +0200 |
User-agent: |
EMH/1.14.1 SEMI/1.14.3 (Ushinoya) FLIM/1.14.2 (Yagi-Nishiguchi) APEL/10.3 Emacs/20.7 (sparc-sun-solaris2.7) (with unibyte mode) |
Inspired by Trevor's regression views, I've done some revisions of the
regression tests.
/Gunnar
Index: regression/connection.tst
===================================================================
RCS file: /cvsroot/gnugo/gnugo/regression/connection.tst,v
retrieving revision 1.4
diff -u -r1.4 connection.tst
--- regression/connection.tst 2001/10/08 02:32:21 1.4
+++ regression/connection.tst 2001/10/17 19:59:39
@@ -73,19 +73,11 @@
19 same_dragon B15 E16
#? [0]*
+loadsgf games/nicklas/nicklas12.sgf 69
+20 connect K17 N17
+#? [1 K18]*
+
# Report number of nodes visited by the tactical reading
10000 get_reading_node_counter
#? [0]&
-
-
-
-
-
-
-
-
-
-
-
-
Index: regression/nicklas2.tst
===================================================================
RCS file: /cvsroot/gnugo/gnugo/regression/nicklas2.tst,v
retrieving revision 1.10
diff -u -r1.10 nicklas2.tst
--- regression/nicklas2.tst 2001/10/08 02:32:21 1.10
+++ regression/nicklas2.tst 2001/10/17 19:59:39
@@ -109,6 +109,9 @@
#DESCRIPTION=Protect those borders!
#SEVERITY=8
#Added A8, for the faint of heart.
+# It's black to move!. The business is to kill white, which is most
+# solidly done with C9, but I think A5 and maybe some other moves
+# work as well. /gf
loadsgf games/nicklas/nicklas17.sgf 47
1701 gg_genmove black
#? [C9|A8]*
Index: regression/nicklas4.tst
===================================================================
RCS file: /cvsroot/gnugo/gnugo/regression/nicklas4.tst,v
retrieving revision 1.3
diff -u -r1.3 nicklas4.tst
--- regression/nicklas4.tst 2001/09/06 21:34:20 1.3
+++ regression/nicklas4.tst 2001/10/17 19:59:39
@@ -19,9 +19,12 @@
809 gg_genmove black
#? [B10]*
+# Changed correct answer from K12 to J12. /gf
+# K14 also secures a connection of all important stones but is one
+# point worse than J12. K12 loses the tail.
loadsgf games/nicklas/nicklas8.sgf 242
812 gg_genmove black
-#? [K12]*
+#? [J12]*
loadsgf games/nicklas/nicklas11.sgf 56
1102 gg_genmove black
@@ -31,6 +34,7 @@
1103 gg_genmove black
#? [C3]*
+# P9 must be an error. /gf
loadsgf games/nicklas/nicklas12.sgf 29
1201 gg_genmove black
#? [P9]*
@@ -39,8 +43,9 @@
1205 gg_genmove black
#? [E16]*
+loadsgf games/nicklas/nicklas12.sgf 215
1215 dragon_status A4
-#? [dead]*
+#? [critical]*
loadsgf games/nicklas/nicklas15.sgf 185
1501 gg_genmove black
Index: regression/nicklas5.tst
===================================================================
RCS file: /cvsroot/gnugo/gnugo/regression/nicklas5.tst,v
retrieving revision 1.4
diff -u -r1.4 nicklas5.tst
--- regression/nicklas5.tst 2001/10/15 00:21:16 1.4
+++ regression/nicklas5.tst 2001/10/17 19:59:39
@@ -31,7 +31,7 @@
loadsgf games/nicklas/nicklas12.sgf 71
1203 gg_genmove black
-#? [N18]*
+#? [N18|N19|M19]*
loadsgf games/nicklas/nicklas12.sgf 89
1204 gg_genmove black
Index: regression/owl.tst
===================================================================
RCS file: /cvsroot/gnugo/gnugo/regression/owl.tst,v
retrieving revision 1.12
diff -u -r1.12 owl.tst
--- regression/owl.tst 2001/10/09 17:56:17 1.12
+++ regression/owl.tst 2001/10/17 19:59:39
@@ -189,7 +189,7 @@
# incident 199
loadsgf games/incident199.sgf 42
54 owl_attack F4
-#? [1 H2]
+#? [1 (H2|J2)]
55 owl_defend F4
#? [1 (H2|G2)]
Index: regression/strategy4.tst
===================================================================
RCS file: /cvsroot/gnugo/gnugo/regression/strategy4.tst,v
retrieving revision 1.12
diff -u -r1.12 strategy4.tst
--- regression/strategy4.tst 2001/10/17 00:07:11 1.12
+++ regression/strategy4.tst 2001/10/17 19:59:40
@@ -177,7 +177,7 @@
# O8 only move.
loadsgf games/owl29.sgf 134
186 gg_genmove white
-#? [1 O8]*
+#? [O8]*
# B can play at C4 if we remember the tesuji B:C4 W:E4 B:B2.
@@ -233,11 +233,11 @@
198 gg_genmove black
#? [C10]
-loadsgf games/bretz 164
+loadsgf games/bretz.sgf 164
199 gg_genmove black
#? [N5]*
-loadsgf games/bretz.sgf 149
+loadsgf games/bretz.sgf 212
200 gg_genmove black
#? [P6|P7|Q7]*
Index: regression/trevor.tst
===================================================================
RCS file: /cvsroot/gnugo/gnugo/regression/trevor.tst,v
retrieving revision 1.15
diff -u -r1.15 trevor.tst
--- regression/trevor.tst 2001/10/17 17:38:29 1.15
+++ regression/trevor.tst 2001/10/17 19:59:40
@@ -46,7 +46,7 @@
4 gg_genmove white
#? [G1]
-#E5 is the only move. Is GnuGo playing somewhat randomly here?
+#E5 is the only move. Is GNU Go playing somewhat randomly here?
#Dan said:
#Looking at unsorted test 5, a typical way to solve this would be
#to add an edge Joseki pattern, such as this:
@@ -433,10 +433,10 @@
#? [C4]*
#CATEGORY=KO
-#DESCRIPTION=Terrible confusion on GnuGo's part here.
+#DESCRIPTION=Terrible confusion on GNU Go's part here.
#SEVERITY=10
#Submitted by:address@hidden (Cai Qiang)
-#Cai said: I made a play between GnuGo 3.0.0 and FunGo with 3
+#Cai said: I made a play between GNU Go 3.0.0 and FunGo with 3
# handicaps. Gnugo played quite well. And in the end it almost
# killed a corner of FunGo, but it played a big blunder:-(.
# Attachment is the SGF file.
Index: regression/vie.tst
===================================================================
RCS file: /cvsroot/gnugo/gnugo/regression/vie.tst,v
retrieving revision 1.3
diff -u -r1.3 vie.tst
--- regression/vie.tst 2001/10/17 00:07:11 1.3
+++ regression/vie.tst 2001/10/17 19:59:40
@@ -74,9 +74,10 @@
18 owl_attack E2
#? [1 F2]
+# S8, T8, T9 and T7 all work. S9 and T10 do not.
loadsgf golois/Goemate990911-3.sgf
19 owl_attack R8
-#? [1 S8]*
+#? [1 S8|T8|T9|T7]*
# B can tenuki. J8 and H9 are miai.
loadsgf golois/Goemate991106-3.sgf
@@ -156,9 +157,10 @@
38 owl_defend B8
#? [1 (B6|A4)]*
+# D7 is one point smaller, but also lives.
loadsgf golois/MeilleurCoup0004.sgf
39 owl_defend B6
-#? [1 C6]*
+#? [1 (C6|D7)]*
loadsgf golois/MeilleurCoup0005.sgf
40 owl_defend D9