[computer-go] More statistics and conclusions from CGOS data
Gunnar Farnebäck
gunnar at lysator.liu.se
Wed Feb 13 15:12:44 PST 2008
I wrote:
> Interesting. If I do the same with MonteGNU's fuseki database, which
> is based on online learning from own CGOS games, and cut off at 200
> samples I get:
>
> [...]
Since I started to prepare the tree with a cut off of 100 samples I
may as well post that too in case someone is interested.
E5 8101
| C3 2950
| | G5 1798
| | | G3 1145 (A)
| | | | C5 403
| | | | | D7 127
| | | | C6 189
| | | | E3 138
| | | C7 310
| | | | C5 111
| | | | | F4 100
| | | | C4 103
| | | | | D4 102
| | | G7 121
| | E3 622
| | | C7 1145 (transposition of A)
| | | C5 158 (B)
| | | G6 138
| | D3 157
| | | F7 150
| | | | D6 111
| | F3 154
| | D4 100
| | | D3 125
| C4 2249
| | C5 714
| | | B5 459
| | | | C6 403
| | | | | D4 128
| | | | | G6 110
| | | | | | G4 125
| | | | | | | E3 115
| | | | | G4 100
| | D4 443
| | | C5 277 (C)
| | | | C3 162
| | | | | B3 117
| | | C3 125
| | C6 351
| | | G6 132
| | D6 286
| | E3 239
| | D3 148
| C5 1708
| | D4 612
| | | C4 277 (transposition of C)
| | | F7 166
| | | | F6 110
| | C4 343
| | | D4 172
| | | | D5 177 (D)
| | | | | C3 167
| | | | | | B4 166
| | | B4 110
| | C3 297
| | D3 172
| | E3 142
| | | C3 158 (transposition of B)
| D4 864
| | E4 698
| | | D5 205 (E)
| | | | D6 159
| | | | | C6 123
| | | E3 138
| | | | D3 177 (transposition of D)
| | | | F3 119
| | | D3 129
| D5 295
| | D4 216
| | E4 114
| | | D4 205 (transposition of E)
D4 1608
| E5 681
| | E4 520 (F)
| | | D5 351
| | | | C5 212
| | | | | C6 154
| | | | | | F5 153
| | | | | | | C4 122
| | | | | | | | B5 117
| | | | F5 171 (G)
| | | | | F6 132
| | | F5 125
| F6 279
| F5 250
| G6 111
D5 1521
| F5 508
| | F4 190
| | | G4 176
| | | | E4 109
| | E4 122
| | | E5 104 (H)
| | | | F4 171 (transposition of G)
| F4 320
| | F5 117
| | F6 114
| E5 181
| | D4 520 (transposition of F)
| | E4 160
| | | F5 104 (transposition of H)
| G4 164
C4 585
| D5 174
| F6 123
| E5 105
C5 391
| F5 135
C3 149
| D4 122
/Gunnar
More information about the computer-go
mailing list