Chess2u
Would you like to react to this message? Create an account in a few clicks or log in to continue.

Chess2uLog in

stockfish of the neural network

descriptionstockfish of the neural network - Page 10 EmptyRe: stockfish of the neural network

more_horiz
BrainLearn 13
CorChess
Eman v7.21
Fisherov 0.98 NNUE
Honey v13.2
Raubfisch GTZ24b
ShashChess 17.1
Stockfish Development
Stockfish exp56 modern Sopel97
SugaR AI 2.10

So far the abv 10 chess engines r supporting the new NNUE network.

descriptionstockfish of the neural network - Page 10 EmptyRe: stockfish of the neural network

more_horiz
Don't forget FF 2 ( Fat Fritz 2 )

descriptionstockfish of the neural network - Page 10 EmptyRe: stockfish of the neural network

more_horiz
Author: bmc4
Date: Sun Jun 6 21:22:39 2021 +0200
Timestamp: 1623007359

Reduce in LMR reduction on PvNode

reduce reduction in LMR by 1 on PvNode.

STC:
LLR: 2.93 (-2.94,2.94) <-0.50,2.50>
Total: 266080 W: 22438 L: 21996 D: 221646 Elo +0.58
Ptnml(0-2): 774, 17874, 95376, 18168, 848
https://tests.stockfishchess.org/tests/view/60bc0661457376eb8bcaa4bb

LTC:
LLR: 2.93 (-2.94,2.94) <0.50,3.50>
Total: 20144 W: 698 L: 587 D: 18859 Elo +1.91
Ptnml(0-2): 2, 529, 8906, 626, 9
https://tests.stockfishchess.org/tests/view/60bcc3f2457376eb8bcaa58d

closes https://github.com/official-stockfish/Stockfish/pull/3534

bench: 5173012

https://abrok.eu/stockfish/

descriptionstockfish of the neural network - Page 10 EmptyRe: stockfish of the neural network

more_horiz
Author: bmc4
Date: Tue Jun 8 20:16:20 2021 +0200
Timestamp: 1623176180

Simplify promotion move generator

This patch removes Knight promotion checks from Captures. As a consequence,
it also removes this underpromotion from qsearch.

STC:
LLR: 2.93 (-2.94,2.94) <-2.50,0.50>
Total: 37776 W: 3113 L: 3023 D: 31640 Elo +0.83
Ptnml(0-2): 103, 2419, 13755, 2507, 104
https://tests.stockfishchess.org/tests/view/60be6a06457376eb8bcaa775

LTC:
LLR: 2.93 (-2.94,2.94) <-2.50,0.50>
Total: 39760 W: 1257 L: 1203 D: 37300 Elo +0.47
Ptnml(0-2): 11, 1079, 17646, 1133, 11
https://tests.stockfishchess.org/tests/view/60beb972457376eb8bcaa7c5

closes https://github.com/official-stockfish/Stockfish/pull/3536

Bench: 5530620

https://abrok.eu/stockfish/

descriptionstockfish of the neural network - Page 10 EmptyRe: stockfish of the neural network

more_horiz
@Zippo wrote:
Don't forget FF 2 ( Fat Fritz 2 )


FF2 cant really make it..

64.5-35.5 this is the best FF2 can do agst Fisherov 0.98 NNUE which is the GOAT now..

https://drive.google.com/file/d/1T1fX5k9NkQn3beJo1C_8-_tvuK-Hhxtq/view?usp=sharing

U can download the PGN file above, Fisherov gave FF2 a good thrashing (FF2 won 5 games though).

descriptionstockfish of the neural network - Page 10 EmptyRe: stockfish of the neural network

more_horiz
The topic was about engines with NEW 45mb NNUE NET.. it didn't talked it's performances nor the strength.. 😉

descriptionstockfish of the neural network - Page 10 EmptyRe: stockfish of the neural network

more_horiz
@Triton6654 wrote:
@Zippo wrote:
Don't forget FF 2 ( Fat Fritz 2 )


FF2 cant really make it..

64.5-35.5 this is the best FF2 can do agst Fisherov 0.98 NNUE which is the GOAT now..

https://drive.google.com/file/d/1T1fX5k9NkQn3beJo1C_8-_tvuK-Hhxtq/view?usp=sharing

U can download the PGN file above, Fisherov gave FF2 a good thrashing (FF2 won 5 games though).


And does anyone have a link to the actual Fisherov engine? Why do they require you to send personal information for a "password"?

descriptionstockfish of the neural network - Page 10 EmptyRe: stockfish of the neural network

more_horiz
Author: Joost VandeVondele
Date: Fri Jun 11 20:33:24 2021 +0200
Timestamp: 1623436404
Limit double extensions

https://abrok.eu/stockfish/

descriptionstockfish of the neural network - Page 10 EmptyRe: stockfish of the neural network

more_horiz
Author: Tomasz Sobczyk
Date: Sun Jun 13 13:10:56 2021 +0200

Author: Vizvezdenec
Date: Sun Jun 13 12:00:20 2021 +0200

Author: Stéphane Nicolet
Date: Sun Jun 13 10:02:43 2021 +0200

Author: Tomasz Sobczyk
Date: Sun Jun 13 09:39:03 2021 +0200

https://abrok.eu/stockfish/

descriptionstockfish of the neural network - Page 10 EmptyRe: stockfish of the neural network

more_horiz
Author: Tomasz Sobczyk
Date: Mon Jun 14 11:22:08 2021 +0200

Timestamp: 1623662528

Reduce the number of accumulator states

Reduce from 3 to 2. Make the intent of the states clearer.

STC: https://tests.stockfishchess.org/tests/view/60c50111457376eb8bcaad03
LLR: 2.95 (-2.94,2.94) <-2.50,0.50>
Total: 61888 W: 5007 L: 4944 D: 51937 Elo +0.35
Ptnml(0-2): 164, 3947, 22649, 4030, 154

LTC: https://tests.stockfishchess.org/tests/view/60c52b1c457376eb8bcaad2c
LLR: 2.94 (-2.94,2.94) <-2.50,0.50>
Total: 20248 W: 688 L: 618 D: 18942 Elo +1.20
Ptnml(0-2): 7, 551, 8946, 605, 15

closes https://github.com/official-stockfish/Stockfish/pull/3548

No functional change.

Author: JWmer
Date: Mon Jun 14 09:24:07 2021 +0200

Timestamp: 1623655447

Update default net to nn-8e47cf062333.nnue

This net is the result of training on data used by the Leela project. More precisely,
we shuffled T60 and T74 data kindly provided by borg (for different Tnn, the data is
a result of Leela selfplay with differently sized Leela nets).

The data is available at vondele's google drive:
https://drive.google.com/drive/folders/1mftuzYdl9o6tBaceR3d_VBQIrgKJsFpl.

The Leela data comes in small chunks of .binpack files. To shuffle them, we simply
used a small python script to randomly rename the files, and then concatenated them
using `cat`. As validation data we picked a file of T60 data. We will further investigate
T74 data.

The training for the NNUE architecture used 200 epochs with the Python trainer from
the Stockfish project. Unlike the previous run we tried with this data, this run does
not have adjusted scaling — not because we didn't want to, but because we forgot.
However, this training randomly skips 40% more positions than previous run. The loss
was very spiky and decreased slower than it does usually.

Training loss: https://github.com/official-stockfish/images/blob/main/training-loss-8e47cf062333.png
Validation loss: https://github.com/official-stockfish/images/blob/main/validation-loss-8e47cf062333.png

This is the exact training command:
python train.py --smart-fen-skipping --random-fen-skipping 14 --batch-size 16384 --threads 4 --num-workers 4 --gpus 1 trainingdata\training_data.binpack validationdata\val.binpack

---

10k STC result:
ELO: 3.61 +-3.3 (95%) LOS: 98.4%
Total: 10000 W: 1241 L: 1137 D: 7622 Elo +3.61
Ptnml(0-2): 68, 841, 3086, 929, 76
https://tests.stockfishchess.org/tests/view/60c67e50457376eb8bcaae70

10k LTC result:
ELO: 2.71 +-2.4 (95%) LOS: 98.8%
Total: 10000 W: 659 L: 581 D: 8760 Elo +2.71
Ptnml(0-2): 22, 485, 3900, 579, 14
https://tests.stockfishchess.org/tests/view/60c69deb457376eb8bcaae98

Passed LTC:
LLR: 2.93 (-2.94,2.94) <0.50,3.50>
Total: 9648 W: 685 L: 545 D: 8418 Elo +5.04
Ptnml(0-2): 22, 448, 3740, 596, 18
https://tests.stockfishchess.org/tests/view/60c6d41c457376eb8bcaaecf

---

closes https://github.com/official-stockfish/Stockfish/pull/3550

Bench: 4877339

https://abrok.eu/stockfish/

descriptionstockfish of the neural network - Page 10 EmptyRe: stockfish of the neural network

more_horiz
https://m.nextchessmove.com/dev-builds

400 elo

descriptionstockfish of the neural network - Page 10 EmptyRe: stockfish of the neural network

more_horiz
Author: J. Oster
Date: Mon Jun 14 17:28:30 2021 +0200
Timestamp: 1623684510

Fix a rare case of wrong TB ranking
of a root move leading to a 3-fold repetition.
With this small fix a draw ranking and thus a draw score is being applied.
This works for both, ranking by dtz or wdl tables.

Fixes https://github.com/official-stockfish/Stockfish/issues/3542

(No functional change without TBs.)
Bench: 4877339

descriptionstockfish of the neural network - Page 10 EmptyRe: stockfish of the neural network

more_horiz
Author: Joost VandeVondele
Date: Tue Jun 15 22:08:40 2021 +0200
Timestamp: 1623787720

New default net nn-33c9d39e5eb6.nnue

As the previous net, this net is trained on Leela games as provided by borg.
See also https://lczero.org/blog/2021/06/the-importance-of-open-data/

The particular data set, which is a mix of T60 and T74 data, is now available as a single binpack:
https://drive.google.com/file/d/1RFkQES3DpsiJqsOtUshENtzPfFgUmEff/view?usp=sharing

The training command was:
python train.py ../../training_data_pylon.binpack ../../training_data_pylon.binpack --gpus 1 --threads 2 --num-workers 2 --batch-size 16384 --progress_bar_refresh_rate 300 --smart-fen-skipping --random-fen-skipping 10 --features=HalfKAv2^ --lambda=1.0 --max_epochs=440 --seed $RANDOM --default_root_dir exp/run_2

passed STC:
https://tests.stockfishchess.org/tests/view/60c887cb457376eb8bcab054
LLR: 2.94 (-2.94,2.94) <-0.50,2.50>
Total: 12792 W: 1483 L: 1311 D: 9998 Elo +4.67
Ptnml(0-2): 62, 989, 4131, 1143, 71

passed LTC:
https://tests.stockfishchess.org/tests/view/60c8e5c4457376eb8bcab0f0
LLR: 2.95 (-2.94,2.94) <0.50,3.50>
Total: 11272 W: 601 L: 477 D: 10194 Elo +3.82
Ptnml(0-2): 9, 421, 4657, 535, 14

also had strong LTC performance against another strong net of the series:
https://tests.stockfishchess.org/tests/view/60c8c40d457376eb8bcab0c6

closes https://github.com/official-stockfish/Stockfish/pull/3557

Bench: 5032320

https://abrok.eu/stockfish/

descriptionstockfish of the neural network - Page 10 EmptyRe: stockfish of the neural network

more_horiz
Author: bmc4
Date: Thu Jun 17 02:08:33 2021 +0200
Timestamp: 1623888513

Author: proukornew
Date: Thu Jun 17 01:14:20 2021 +0200
Timestamp: 1623885260


https://abrok.eu/stockfish/

descriptionstockfish of the neural network - Page 10 EmptyRe: stockfish of the neural network

more_horiz
Very interesting that Stockfish neural nets are starting to use Leela data. This is a pretty big development.

descriptionstockfish of the neural network - Page 10 EmptyRe: stockfish of the neural network

more_horiz
Author: Stéphane Nicolet
Date: Thu Jun 17 18:10:01 2021 +0200
Timestamp: 1623946201

Revert "Fix for Cygwin's environment build-profile"

This reverts commit "Fix for Cygwin's environment build-profile", as it was
giving errors for "make clean" on some Windows environments. See comments in
https://github.com/official-stockfish/Stockfish/commit/68bf362ea2385a641be9f5ed9ce2acdf55a1ecf1

Possibly somebody can propose a solution that would fix Cygwin builds and
not break on other system too, stay tuned! :-)

No functional change

https://abrok.eu/stockfish/

descriptionstockfish of the neural network - Page 10 EmptyRe: stockfish of the neural network

more_horiz
Permissions in this forum:
You cannot reply to topics in this forum