Life In 19x19 http://lifein19x19.com/ |
|
KataGo Distributed Training and new networks http://lifein19x19.com/viewtopic.php?f=18&t=17944 |
Page 15 of 23 |
Author: | hope366 [ Fri Nov 25, 2022 4:02 pm ] |
Post subject: | Re: KataGo Distributed Training and new networks |
If anyone has a CUDA version of KataGo for the new architecture, could you please upload it? I tried to compile using cmake before, but it didn't work in my environment. |
Author: | And [ Fri Nov 25, 2022 4:16 pm ] |
Post subject: | Re: KataGo Distributed Training and new networks |
and the version of eigenavx2 for Windows please! |
Author: | And [ Fri Nov 25, 2022 10:24 pm ] | ||
Post subject: | Re: KataGo Distributed Training and new networks | ||
KataGo 18b, playouts 10 - 60b, playouts 10
|
Author: | And [ Sat Nov 26, 2022 4:43 am ] |
Post subject: | Re: KataGo Distributed Training and new networks |
PC_Screen wrote: Katago Eigen, same process, drop it inside the katago 1.11 eigen folder and replace the old katago.exe. https://drive.google.com/file/d/18XJHc8 ... sp=sharing As for the speed being lower, try running the benchmark in the command prompt (./katago.exe benchmark -model <NEURALNET>.bin.gz) and adjust the number of search threads based on it many thanks! it works, with different networks, and 18b works. but the speed compared to eigen, eigenavx2 1.11.0 is tens times slower. something's wrong. visits: 1 10b eigen 1.11 Time taken: 0.281 10b your file Time taken: 15.21 60b eigen 1.11 Time taken: 2.605 60b your file Time taken: 160.166 18b your file Time taken: 56.941 |
Author: | And [ Sat Nov 26, 2022 10:05 am ] | ||
Post subject: | Re: KataGo Distributed Training and new networks | ||
second game KataGo 18b, playouts 10 - 60b, playouts 10 (60b-s665)
|
Author: | And [ Sat Nov 26, 2022 8:13 pm ] | ||
Post subject: | Re: KataGo Distributed Training and new networks | ||
KataGo 18b, playouts 100 - CS Zero 9d H4
|
Author: | And [ Sun Nov 27, 2022 8:16 am ] | ||
Post subject: | Re: KataGo Distributed Training and new networks | ||
KataGo 18b, playouts 100 - CS Zero 9d H5 after 347 moves CS Zero: winrate 64.3, B+32
|
Author: | PC_Screen [ Sun Nov 27, 2022 6:23 pm ] | ||
Post subject: | Re: KataGo Distributed Training and new networks | ||
New KataGo 18b out, supposed to as strong as 60b at equal visits while also being faster than 40b, which means it should be stronger than both in an actual game. https://cdn.discordapp.com/attachments/ ... 21b.bin.gz Also test game: KataGo 1.11 18b (W) vs KataGo 1.3.5 40b (B) with 2H and 0.5 Komi, Japanese Rules. The time setting is 10s/move for both which gave them around 1-2k visits per move on average. The new 18b KataGo managed a convincing win against a 2 year old KataGo despite the 2 handicap stones.
|
Author: | And [ Sun Nov 27, 2022 6:27 pm ] |
Post subject: | Re: KataGo Distributed Training and new networks |
"UEC finished this weekend...KataGo placed 2nd in the final (and first in the preliminaries)." https://discord.com/channels/4170221623 ... 4930181303 |
Author: | And [ Mon Nov 28, 2022 12:47 pm ] | ||
Post subject: | Re: KataGo Distributed Training and new networks | ||
KataGo (18b-uec, playouts 1) - CS Zero 9d
|
Author: | And [ Mon Nov 28, 2022 9:37 pm ] |
Post subject: | Re: KataGo Distributed Training and new networks |
100 games 19x19 KataGo 18b-uec (playouts 1) - 60b-s665 (playouts 1): 18b (B) - 60b (W) 21:29 18b (W) - 60b (B) 26:24 |
Author: | And [ Tue Nov 29, 2022 12:05 am ] |
Post subject: | Re: KataGo Distributed Training and new networks |
100 games 19x19 KataGo 18b-uec (playouts 2) - 60b-s665 (playouts 1): 18b (B) - 60b (W) 32:18 18b (W) - 60b (B) 35:15 |
Author: | And [ Tue Nov 29, 2022 6:32 am ] | ||
Post subject: | Re: KataGo Distributed Training and new networks | ||
10 games KataGo 18b-uec (playouts 10) - 60b-s665 (playouts 5): 18b (W) - 60b (B) 8:2
|
Author: | And [ Tue Nov 29, 2022 10:52 am ] | ||
Post subject: | Re: KataGo Distributed Training and new networks | ||
10 games KataGo 18b-uec (playouts 10) - 60b-s665 (playouts 5): 18b (B) - 60b (W) 9:1
|
Author: | And [ Tue Nov 29, 2022 11:28 am ] |
Post subject: | Re: KataGo Distributed Training and new networks |
in the position from here https://lifein19x19.com/viewtopic.php?p=244851#p244851, I turned on the analysis mode, the network 18b-uec. the move h1 was immediately highlighted. after ~200 visits, katago showed white winning, but then decided that there no winning and the winrate continued to fall EDIT after 10K visits - 39.4 |
Author: | PC_Screen [ Tue Nov 29, 2022 8:45 pm ] | ||||
Post subject: | Re: KataGo Distributed Training and new networks | ||||
Result of a 100 games match between the latest 18b vs the strongest 40b (S1184), 1s/move (averaging 200-500 visits/move), Chinese Rules and komi is set to 7.0 which allows for draws. A draw is counted as +0.5 points for both players: 18b vs 40b: 65 : 35 From 18b's perspective: Wins: 58 Draws: 14 Losses: 28 Elo Difference: +126.51 ± 73.59
|
Author: | And [ Wed Nov 30, 2022 5:00 am ] | ||
Post subject: | Re: KataGo Distributed Training and new networks | ||
10 games KataGo 18b-uec (playouts 20) - 60b-s665 (playouts 10): 18b (B) - 60b (W) 9:1
|
Author: | And [ Fri Dec 02, 2022 10:21 am ] | ||
Post subject: | Re: KataGo Distributed Training and new networks | ||
10 games KataGo 18b-uec (playouts 20) - 60b-s665 (playouts 10): 18b (W) - 60b (B) 10:0
|
Author: | PC_Screen [ Sat Dec 03, 2022 9:25 am ] |
Post subject: | Re: KataGo Distributed Training and new networks |
Thorough test done by @Medwin on the Go AI Discord, using 400 visits/move with a total of 2800 games played. Attachment: image.png [ 350.38 KiB | Viewed 13383 times ] As expected, the new b18-uec is slightly stronger than 60b at equal visits (much stronger with equal time) and is expected to grow even stronger as its training run progresses. When KataGo 1.12 releases 18b is expected to replace 60b as the main network for training games. |
Author: | PC_Screen [ Tue Dec 06, 2022 2:26 pm ] | |||
Post subject: | Re: KataGo Distributed Training and new networks | |||
Result of a 100 games match between the latest 18b vs the strongest 20b, 400 visits/move, Chinese Rules and komi is set to 7.0 which allows for draws. A draw is counted as +0.5 points for both players: 18b vs 20b: 92.5 : 7.5 From 18b's perspective: Wins: 92 Draws: 1 Losses: 7 Elo Difference: +447.48 ± 132.32 Also did the same against Leela Zero, 1s/move, Tromp-Taylor Rules, 7.5 Komi, and the result was: 18b vs LZ 40B: 100:0 Elo can't be calculated from this since 18b didn't lose any games. Most estimates put Leela ~800 elo or so behind KataGo Example Games: 18b vs 20b 18b vs LZ 40b:
|
Page 15 of 23 | All times are UTC - 8 hours [ DST ] |
Powered by phpBB © 2000, 2002, 2005, 2007 phpBB Group http://www.phpbb.com/ |