LC0 finally?
Moderators: hgm, Harvey Williamson, bob
Forum rules
This textbox is used to restore diagrams posted with the [d] tag before the upgrade.
This textbox is used to restore diagrams posted with the [d] tag before the upgrade.
LC0 finally?
This looks really promising with the latest networks, is it as good as it looks?
Re: LC0 finally?
No. It's not bad generally for 11th net (ID4011), but the rating is misleading. ID4011 is some 500 Elo points weaker than the best nets of the run 1 and run 5 against AB engines, but shows some 400 Elo points better rating in self-games than the best nets of those runs, which is misleading. The rating of the latest run are very inflated, but the real progress against AB engines is still not bad for only 11th net.
Re: LC0 finally?
4018 reads at 3400 elo, so is that about 2900-3000 in real terms?
Not bad....
Not bad....
Re: LC0 finally?
About 2900 in CCRL 40/4' conditions on GTX 1060 GPU. ID4018 is about the level of ID1028 of the previous run (or shown 2500 Elo level of that run), which means a significantly faster progress (18 nets compared to 28 nets, each needing about 40,000 games). The shown rating in self-games now is very inflated, for obscure to me reasons.
Re: LC0 finally?
I think it's just id4000 vs id4001 match which gave very low precision Elo delta with just 1 lost game (result was +423 -1 =0).Laskos wrote: ↑Thu Jun 21, 2018 9:44 amAbout 2900 in CCRL 40/4' conditions on GTX 1060 GPU. ID4018 is about the level of ID1028 of the previous run (or shown 2500 Elo level of that run), which means a significantly faster progress (18 nets compared to 28 nets, each needing about 40,000 games). The shown rating in self-games now is very inflated, for obscure to me reasons.
If it was 2 lost games instead of 1, it would be 400 points lower. And if it was 0 lost games instead of 1, it would be +inf points higher.
-
- Posts: 1757
- Joined: Tue Jun 02, 2009 10:14 pm
Re: LC0 finally?
is the testserver about ready to switch to a larger net then? think on the first one it was quite quickly; i suppose the thinking is too quickly.
Re: LC0 finally?
Update -> http://rebel13.nl/zero.html
After the w396 bootstrap lczero (w429) is climbing and climbing again.
Note that the NPS is 33 only.
After the w396 bootstrap lczero (w429) is climbing and climbing again.
Note that the NPS is 33 only.
90% of coding is debugging, the other 10% is writing bugs.
Re: LC0 finally?
By ID4039, in my gauntlet at fast TC against AB engines, it's already similar in strength to best 6x64 nets of the previous runs, scoring 61.0/200. Probably close to saturation, but let's see (if they won't start from scratch again).
Re: LC0 finally?
Some people claim that version 0604-experimental plays stronger than 0619 on the same net. That may theoretically happen both because algorithm differences (I changed some formulas to be able to save some memory and floats may be rounded differently sometimes) or maybe there is nps difference.
I wasn't able to find any differences in my tests though.
Maybe someone has time to do tests and confirm/refute that claim? That would be much appreciated!
I wasn't able to find any differences in my tests though.
Maybe someone has time to do tests and confirm/refute that claim? That would be much appreciated!
Re: LC0 finally?
Leela seems close to make progress again.