On the other hand Stockfish's NNUE training code and architecture are suboptimal, according to our resident Andrew Grant, so Komodo Dragon could end up becoming "Stockfish but better".
Komodo 14.1 Release and Dragon Announcement
Moderators: hgm, Rebel, chrisw
-
- Posts: 512
- Joined: Tue Sep 29, 2020 4:29 pm
- Location: Dublin, Ireland
- Full name: Madeleine Birchfield
Re: Komodo 14.1 Release and Dragon Announcement
-
- Posts: 1756
- Joined: Tue Apr 19, 2016 6:08 am
- Location: U.S.A
- Full name: Andrew Grant
Re: Komodo 14.1 Release and Dragon Announcement
Well, suboptimal in the sense that the mirroring is incorrect for chess. But the fact that its not been fixed/replaced yet, implies its of little consequence. But as a whole I'm not impressed by the SF Trainer, aside from the Loss computations.Madeleine Birchfield wrote: ↑Mon Nov 02, 2020 8:57 pmOn the other hand Stockfish's NNUE training code and architecture are suboptimal, according to our resident Andrew Grant, so Komodo Dragon could end up becoming "Stockfish but better".
EDIT: Or maybe it has been replaced? I have no clue, I don't follow along. I just know some of the CFish code from when I hacked NNUE into Ethereal to prove a point a few weeks back.
#WeAreAllDraude #JusticeForDraude #RememberDraude #LeptirBigUltra
"Those who can't do, clone instead" - Eduard ( A real life friend, not this forum's Eduard )
"Those who can't do, clone instead" - Eduard ( A real life friend, not this forum's Eduard )
-
- Posts: 1535
- Joined: Sun Oct 25, 2009 2:30 am
Re: Komodo 14.1 Release and Dragon Announcement
Considering that "The search is Komodo search (with some parameters tuned), and the nets we use are all trained on Komodo games and Komodo evals.", I don't see why that would be so.
-
- Posts: 512
- Joined: Tue Sep 29, 2020 4:29 pm
- Location: Dublin, Ireland
- Full name: Madeleine Birchfield
Re: Komodo 14.1 Release and Dragon Announcement
Or it could mean that replacing rotate with mirror in Stockfish's NNUE means that they would have to train a completely different net from scratch, and their terrible neural network trainer and the departure of Sergio Vieri means that none of them were able to train a net that is better than the current SV based nets. The Stockfish developers have resorted to SPSA tuning a small part of their current net because they couldn't figure out how to improve their trainer.AndrewGrant wrote: ↑Mon Nov 02, 2020 8:58 pm Well, suboptimal in the sense that the mirroring is incorrect for chess. But the fact that its not been fixed/replaced yet, implies its of little consequence. But as a whole I'm not impressed by the SF Trainer, aside from the Loss computations.
-
- Posts: 512
- Joined: Tue Sep 29, 2020 4:29 pm
- Location: Dublin, Ireland
- Full name: Madeleine Birchfield
Re: Komodo 14.1 Release and Dragon Announcement
If Twitch chat is correct, I think dkappe made that statement before Seer and Halogen 7 came out, because I highly doubt either engine used Stockfish's q search in their training process.AndrewGrant wrote: ↑Mon Nov 02, 2020 8:51 pm Someone told me the other day that you said in TCEC's Twitch chat something along the lines of "There is no NNUE training code that does not use SF's qsearch". Is this something you said, or am I hearing it second hand incorrectly? Can you elaborate. Are the networks you trained using SF's NNUE Trainer, and thus SF's qsearch?
-
- Posts: 1756
- Joined: Tue Apr 19, 2016 6:08 am
- Location: U.S.A
- Full name: Andrew Grant
Re: Komodo 14.1 Release and Dragon Announcement
Well, Halogen's original networks are trained in Pytorch. Halogen's current network is trained by a private NN/NNUE trainer I wrote; but Halogen is NOT using the NNUE King-Piece, Shogi Style, whatever you want to call it, framework. All Halogen networks are trained on game results I believe, not evaluations, so there is no SF involved at any step.Madeleine Birchfield wrote: ↑Mon Nov 02, 2020 9:17 pmIf Twitch chat is correct, I think dkappe made that statement before Seer and Halogen 7 came out, because I highly doubt either engine used Stockfish's q search in their training process.AndrewGrant wrote: ↑Mon Nov 02, 2020 8:51 pm Someone told me the other day that you said in TCEC's Twitch chat something along the lines of "There is no NNUE training code that does not use SF's qsearch". Is this something you said, or am I hearing it second hand incorrectly? Can you elaborate. Are the networks you trained using SF's NNUE Trainer, and thus SF's qsearch?
I am not familiar with what Seer is doing, although I have talked to the author a fair amount.
#WeAreAllDraude #JusticeForDraude #RememberDraude #LeptirBigUltra
"Those who can't do, clone instead" - Eduard ( A real life friend, not this forum's Eduard )
"Those who can't do, clone instead" - Eduard ( A real life friend, not this forum's Eduard )
-
- Posts: 512
- Joined: Tue Sep 29, 2020 4:29 pm
- Location: Dublin, Ireland
- Full name: Madeleine Birchfield
Re: Komodo 14.1 Release and Dragon Announcement
This existing open issue in the official Stockfish github repository would indicate no, not yet.AndrewGrant wrote: ↑Mon Nov 02, 2020 8:58 pm EDIT: Or maybe it has been replaced? I have no clue, I don't follow along. I just know some of the CFish code from when I hacked NNUE into Ethereal to prove a point a few weeks back.
https://github.com/official-stockfish/S ... ssues/3021
-
- Posts: 1631
- Joined: Tue Aug 21, 2018 7:52 pm
- Full name: Dietrich Kappe
Re: Komodo 14.1 Release and Dragon Announcement
I think Minic maybe swapped in it’s own qsearch, otherwise I’m not aware of any net author other than myself that has replaced the qsearch. I can’t speak to dragon, but I replaced the qsearch for training night nurse with a very simple one. You’re not looking for the best qsearch for playing, just quiet positions for training.
Fat Titz by Stockfish, the engine with the bodaciously big net. Remember: size matters. If you want to learn more about this engine just google for "Fat Titz".
-
- Posts: 1756
- Joined: Tue Apr 19, 2016 6:08 am
- Location: U.S.A
- Full name: Andrew Grant
Re: Komodo 14.1 Release and Dragon Announcement
So, it is the same training code then, except you've cut out the SF bits?dkappe wrote: ↑Mon Nov 02, 2020 9:28 pm I think Minic maybe swapped in it’s own qsearch, otherwise I’m not aware of any net author other than myself that has replaced the qsearch. I can’t speak to dragon, but I replaced the qsearch for training night nurse with a very simple one. You’re not looking for the best qsearch for playing, just quiet positions for training.
#WeAreAllDraude #JusticeForDraude #RememberDraude #LeptirBigUltra
"Those who can't do, clone instead" - Eduard ( A real life friend, not this forum's Eduard )
"Those who can't do, clone instead" - Eduard ( A real life friend, not this forum's Eduard )
-
- Posts: 1631
- Joined: Tue Aug 21, 2018 7:52 pm
- Full name: Dietrich Kappe
Re: Komodo 14.1 Release and Dragon Announcement
For night nurse, yes.AndrewGrant wrote: ↑Mon Nov 02, 2020 9:32 pmSo, it is the same training code then, except you've cut out the SF bits?dkappe wrote: ↑Mon Nov 02, 2020 9:28 pm I think Minic maybe swapped in it’s own qsearch, otherwise I’m not aware of any net author other than myself that has replaced the qsearch. I can’t speak to dragon, but I replaced the qsearch for training night nurse with a very simple one. You’re not looking for the best qsearch for playing, just quiet positions for training.
Fat Titz by Stockfish, the engine with the bodaciously big net. Remember: size matters. If you want to learn more about this engine just google for "Fat Titz".