I'm running my tuner now with the E12.52 set, it shows some values while it is running, the piece values for the opening phase keep dropping, for the endgame they keep rising, this is usually a bad sign. Maybe it doesn't matter, but I don't like it when the mg/eg values get skewed like this.D Sceviour wrote: ↑Fri Oct 16, 2020 2:09 pm I ran a test and got no useful results. That is, the tuning was unable to produce any improvement over Schooners master PST's.The test was compared against my new 1 million epd position filtered training set which I might post. The Ethereal E12 sets above had 80,000 duplicate positions.Code: Select all
1 Schooner2.25-sse 2 8 3000 50% 60% 2 Schooner-E12_2 2 8 3000 50% 61% 3 Schooner1_125_end_2nd 1 8 3000 50% 60% 4 Schooner-E12_1 -4 8 3000 49% 59%
Ethereal Tuning - Data Dump
Moderators: hgm, Rebel, chrisw
-
- Posts: 1563
- Joined: Thu Jul 16, 2009 10:47 am
- Location: Almere, The Netherlands
Re: Ethereal Tuning - Data Dump
-
- Posts: 1563
- Joined: Thu Jul 16, 2009 10:47 am
- Location: Almere, The Netherlands
Re: Ethereal Tuning - Data Dump
Maybe Rofchade uses NNUE, I dont' know. Many engines suddenly have NNUE, probably by just copy-pasting code from Stockfish and adding a precooked network too. Writing optimized code for NNUE and training a network with it is not something you can do in a couple of days, these things take weeks or even months when you have to implement everything from scratch without prior knowledge or experience in this field.chrisw wrote: ↑Fri Oct 16, 2020 2:18 pm it was marked as rofchade, not rpirofchade, so I guess the full on version, 64x. Significant was not really the hardware it was on, but the absolute dominance of the eval function, they were evaluation crushes not search ones. mobility and positional. not fireworks, which I guess figures, those small nets are not going to be highly knowledge packed for difficult king safety/king tactical stuff, but it looks like the knowledge packing/tuning for standard structural features is more than enough.
-
- Posts: 1563
- Joined: Thu Jul 16, 2009 10:47 am
- Location: Almere, The Netherlands
Re: Ethereal Tuning - Data Dump
With the E12.33 set the weights that I get look very similar to the weights I get with my own 4M training set. I wonder what the difference between 12.33 and 12.52 is. I haven't checked the Elo differences yet.Joost Buijs wrote: ↑Fri Oct 16, 2020 6:05 pmI'm running my tuner now with the E12.52 set, it shows some values while it is running, the piece values for the opening phase keep dropping, for the endgame they keep rising, this is usually a bad sign. Maybe it doesn't matter, but I don't like it when the mg/eg values get skewed like this.D Sceviour wrote: ↑Fri Oct 16, 2020 2:09 pm I ran a test and got no useful results. That is, the tuning was unable to produce any improvement over Schooners master PST's.The test was compared against my new 1 million epd position filtered training set which I might post. The Ethereal E12 sets above had 80,000 duplicate positions.Code: Select all
1 Schooner2.25-sse 2 8 3000 50% 60% 2 Schooner-E12_2 2 8 3000 50% 61% 3 Schooner1_125_end_2nd 1 8 3000 50% 60% 4 Schooner-E12_1 -4 8 3000 49% 59%
-
- Posts: 4315
- Joined: Tue Apr 03, 2012 4:28 pm
Re: Ethereal Tuning - Data Dump
Well, those sets zapped my carefully tuned king attack code down to not very much. Noticeable change in play style.Joost Buijs wrote: ↑Sat Oct 17, 2020 11:00 amWith the E12.33 set the weights that I get look very similar to the weights I get with my own 4M training set. I wonder what the difference between 12.33 and 12.52 is. I haven't checked the Elo differences yet.Joost Buijs wrote: ↑Fri Oct 16, 2020 6:05 pmI'm running my tuner now with the E12.52 set, it shows some values while it is running, the piece values for the opening phase keep dropping, for the endgame they keep rising, this is usually a bad sign. Maybe it doesn't matter, but I don't like it when the mg/eg values get skewed like this.D Sceviour wrote: ↑Fri Oct 16, 2020 2:09 pm I ran a test and got no useful results. That is, the tuning was unable to produce any improvement over Schooners master PST's.The test was compared against my new 1 million epd position filtered training set which I might post. The Ethereal E12 sets above had 80,000 duplicate positions.Code: Select all
1 Schooner2.25-sse 2 8 3000 50% 60% 2 Schooner-E12_2 2 8 3000 50% 61% 3 Schooner1_125_end_2nd 1 8 3000 50% 60% 4 Schooner-E12_1 -4 8 3000 49% 59%
-
- Posts: 4366
- Joined: Fri Mar 10, 2006 5:23 am
- Location: http://www.arasanchess.org
Re: Ethereal Tuning - Data Dump
I also go no improvement with the 12.52 training set over my default set. In fact it was a regression. YMMV.
--Jon
--Jon
-
- Posts: 1563
- Joined: Thu Jul 16, 2009 10:47 am
- Location: Almere, The Netherlands
Re: Ethereal Tuning - Data Dump
I've got very strange weights with the E12.52 set, after this I re-tuned with the E12.33 set, now the weights look very similar to the weights I get with my own 4M set. I didn't check performance with self-play yet, but I used these weights during the last blitz tourney. Besides the loss against CoronaVirusChess (due to a questionable position that arose from my opening book) it performed reasonably well.
-
- Posts: 570
- Joined: Mon Jul 20, 2015 5:06 pm
Re: Ethereal Tuning - Data Dump
There are no guarantees that a tuning set can help at arriving at a unique minimum for the generation of PST's. Each test might be converging on a local minimum. There may be at least two major locals. There may be a local minimum for passed pawns, and another one for mating positions or king safety values. Another local might be a combination of these, meaning both are being threatened in a position. The local minimums still have to be defined uniquely for every position using an evaluation function. The evaluator will tell us whether pawn promotions, or mating threats are more important. However, there is no unique regression line that can cover both possibilities. Can NNUE layer design fix this?
-
- Posts: 146
- Joined: Fri Mar 15, 2019 8:46 pm
- Location: Germany
- Full name: Fabian von der Warth
Re: Ethereal Tuning - Data Dump
I've got >= 40 elo out of this data so far.
Author of FabChess: https://github.com/fabianvdW/FabChess
A UCI compliant chess engine written in Rust.
FabChessWiki: https://github.com/fabianvdW/FabChess/wiki
fabianvonderwarth@gmail.com
A UCI compliant chess engine written in Rust.
FabChessWiki: https://github.com/fabianvdW/FabChess/wiki
fabianvonderwarth@gmail.com
-
- Posts: 9
- Joined: Tue Oct 06, 2015 5:00 pm
Re: Ethereal Tuning - Data Dump
What methodology did you use for generating these positions? Specifically, how were the games generated and how did you decide which individual positions to extract from each game? I ask because I would like to take a swing at generating my own positions in the future
Thanks in advance
Thanks in advance
-
- Posts: 1754
- Joined: Tue Apr 19, 2016 6:08 am
- Location: U.S.A
- Full name: Andrew Grant
Re: Ethereal Tuning - Data Dump
The data generation method is outlined in section 2 of this document:jwilson82 wrote: ↑Sun Nov 01, 2020 11:30 pm What methodology did you use for generating these positions? Specifically, how were the games generated and how did you decide which individual positions to extract from each game? I ask because I would like to take a swing at generating my own positions in the future
Thanks in advance
https://github.com/AndyGrant/Ethereal/b ... Tuning.pdf
#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 )