Page 11 of 15
Re: KataGo V1.3
Posted: Sat Mar 14, 2020 7:04 am
by go4thewin
with the ladder code version of leela zero, it's much more even.
https://github.com/yssaya/leela-zero-ladder/releases
katago s243 vs lz 270 ladder version, both bots 1 thread 512 playouts:
lz wins 3 games kata wins 1. Katago lost the first three but won the last easily
Re: KataGo V1.3
Posted: Sat Mar 14, 2020 9:48 am
by lightvector
Some new nets have been uploaded:
Roughly 50 Elo gain for the 20-block extended training net, and slightly more than 100 Elo gain for the 30 and 40-block nets. Enjoy!
https://github.com/lightvector/KataGo/releases
Re: KataGo V1.3
Posted: Sat Mar 14, 2020 10:18 pm
by Vargo
A last 50 game test for the 20b net : g170e-b20c256x2-s2430231552-d525879064
KataGo 1.3.3 g170e-b20c256x2-s2430231552-d525879064 v. LZ017#268 LADDER VERSION
Katago wins 35-15 = 70% (against the "regular" LZ017#268, KG won 74% and 68%)
Twogtp 1.5.1, 1600 visits for both (~ time parity), no error, no duplicate game, all games by resignation.
Details : (KG always appears as W, because of the command -alternate)
LADDER VERSION :
https://github.com/yssaya/leela-zero-ladder/releases
Re: KataGo V1.3
Posted: Sat Mar 14, 2020 11:29 pm
by goame
lightvector wrote:Some new nets have been uploaded:
Roughly 50 Elo gain for the 20-block extended training net, and slightly more than 100 Elo gain for the 30 and 40-block nets. Enjoy!
https://github.com/lightvector/KataGo/releases
Thx for the big nets
b30c
320x2
b
40c256x2
From what I have seen so far, if the second number is bigger (320), then the net has a very good chance to be stronger and if the first number is bigger (40), then the net has a medium chance to be stronger.
Some ideas which could be tried:
30 and 320 vs 30 and 384
30 and 320 vs 30 and 512
40 and 256 vs 40 and 320
40 and 256 vs 40 and 384
40 and 256 vs 40 and 512
Re: KataGo V1.3
Posted: Sun Mar 15, 2020 11:32 am
by Vargo
50 game test between the two newest 20b nets :
g170e-b20c256x2-s2430231552-d525879064 v. g170e-b20c256x2-s2971705856-d633407024
The newest ,
g170e-b20c256x2-s2971705856-d633407024 wins 27-23 (54%)
Twogtp 1.5.1, 1600 visits for both (~ time parity), no error, no duplicate game, all games by resignation.
Details : (s2971 always appears as W, because of the command -alternate)
Re: KataGo V1.3
Posted: Sun Mar 15, 2020 11:36 am
by Vargo
50 game test
KG 1.3.3 g170e-b20c256x2-s2971705856-d633407024 v. LZ#270
KG wins 30-20 (
60%, "only" , maybe 50 games is not enough, or maybe s2971 doesn't like #270
)
Again, Twogtp 1.5.1, 1600 visits for both (~ time parity), no error, no duplicate game, all games by resignation.
Details :
Re: KataGo V1.3
Posted: Sun Mar 15, 2020 1:20 pm
by And
Vargo in this case, maybe another 50?
Re: KataGo V1.3
Posted: Sun Mar 15, 2020 3:01 pm
by go4thewin
katago 1.3.3 20b s297 1 thread 12 playouts vs gtp4zen7 9d setting; 4-0
Wow I can't beieve it is possibe to have a bot this strong and fast on so few playouts.
Re: KataGo V1.3
Posted: Mon Mar 16, 2020 12:17 pm
by Vargo
50 more games
KG 1.3.3 g170e-b20c256x2-s2971705856-d633407024 v. LZ#270
KataGo wins 37-13 = 74% (last test, it was 60%)
details : (KG always appears as W, because of the command -alternate)
Re: KataGo V1.3
Posted: Mon Mar 16, 2020 12:24 pm
by And
LZ will never catch up with the KataGo. Does anyone see the point in the further development of LZ? what are its advantages? different game style?
Re: KataGo V1.3
Posted: Mon Mar 16, 2020 12:36 pm
by And
to try to beat KataGo, you can play with a handicap or with a small playout. in which case does KataGo play more interesting?
Re: KataGo V1.3
Posted: Mon Mar 16, 2020 8:56 pm
by splee99
It depends on your strength, of course. For me, I can't even beat it with 1 playout.
Re: KataGo V1.3
Posted: Tue Mar 17, 2020 4:39 am
by And
in which of the 2 options is the game more interesting?
Re: KataGo V1.3
Posted: Tue Mar 17, 2020 7:15 pm
by go4thewin
I think the most interesting is to play against a bot exactly at your level. Find one from the list below. The ranks do not correspond to human ranks. Spend the money to get zen 6 or 7. In strength, Zen 6 ends right where katago starts, so between gtp4zen and katago, you have the full range of human play, except pure beginner. Sabaki allows for any standard size board, komi, and either common ruleset.
Re: KataGo V1.3
Posted: Tue Mar 17, 2020 8:10 pm
by Bill Spight
go4thewin wrote:I think the most interesting is to play against a bot exactly at your level. Find one from the list below. The ranks do not correspond to human ranks. Spend the money to get zen 6 or 7.
One thing to consider is that humans and neural nets have different strengths and weaknesses. I don't know about today's bots, but kyu level bots before the advent of AlphaGo were bad at life and death and at the endgame, by comparison with humans of similar strength. This relative weakness of the bots showed up in particular at the end of play, when the dame get filled. If the same is true of today's bots, there will be opportunities at the end of play to swing the game, either honestly or be pulling off a swindle.