jirayu_sri/stockfish
1
0
Fork 0
mirror of https://github.com/official-stockfish/Stockfish.git synced 2026-07-22 12:47:08 +00:00
Code Issues Packages Projects Releases Wiki Activity
5,140 Commits 6 Branches 60 Tags
4260ed0c7fec950a14b4cb42fc62f046076f9c0c
Commit Graph
99 Commits
This Branch
This Branch
All Branches
Author SHA1 Message Date
nodchip 4260ed0c7f Merge branch 'master' of github.com:official-stockfish/Stockfish into nnue-player-merge 2020-08-10 08:52:55 +09:00
nodchip 4f97d3446d Cleaned up source code. 2020-08-10 08:52:34 +09:00
nodchip 7f1f08d094 Merge branch 'master' of github.com:official-stockfish/Stockfish into nnue-player-merge 
# Conflicts:
#	README.md
 2020-08-09 09:19:47 +09:00
nodchip fcd70a3c81 Updated README.md. 
Bench: 4067325
 2020-08-08 21:00:19 +09:00
nodchip 3b5de9f18b Merge branch 'master' of github.com:official-stockfish/Stockfish into nnue-player-merge 2020-08-08 19:47:32 +09:00
nodchip 22b85810fe Re-added the code to skip loading a net file. 2020-08-08 19:04:08 +09:00
nodchip 4f94f29f39 Revert "Fixed a bug that the training data generation crashes if eval_limit is high." 
This reverts commit b0d28ac3ab.
 2020-08-08 18:38:02 +09:00
nodchip 9a0b20d3fc Changed to show if NNUE is used in the training data generator. 2020-08-08 18:24:09 +09:00
nodchip b0d28ac3ab Fixed a bug that the training data generation crashes if eval_limit is high. 2020-08-08 18:23:11 +09:00
nodchip 70d88364fe Fixed a bug that the training data generation crashes. 2020-08-08 18:22:29 +09:00
nodchip ed4d007e3c Fixed a bug that the training data generator crahses on memory allocation. 2020-08-08 18:21:38 +09:00
nodchip 2395833c07 Re-added commands for training data generator and trainer. 2020-08-08 16:52:18 +09:00
nodchip fa649ba1e2 Removed a compile warning. 2020-08-08 16:17:55 +09:00
nodchip 2c9075e919 Fixed Makefile to fix build. 2020-08-08 16:05:05 +09:00
nodchip 55a6b2bdc4 Merge branch 'master' of github.com:official-stockfish/Stockfish into nnue-player-merge 
# Conflicts:
#	README.md
#	Readme.md
#	src/Makefile
#	src/evaluate.cpp
#	src/evaluate.h
#	src/misc.cpp
#	src/nnue/architectures/halfkp_256x2-32-32.h
#	src/nnue/evaluate_nnue.cpp
#	src/nnue/evaluate_nnue.h
#	src/nnue/features/feature_set.h
#	src/nnue/features/features_common.h
#	src/nnue/features/half_kp.cpp
#	src/nnue/features/half_kp.h
#	src/nnue/features/index_list.h
#	src/nnue/layers/affine_transform.h
#	src/nnue/layers/clipped_relu.h
#	src/nnue/layers/input_slice.h
#	src/nnue/nnue_accumulator.h
#	src/nnue/nnue_architecture.h
#	src/nnue/nnue_common.h
#	src/nnue/nnue_feature_transformer.h
#	src/position.cpp
#	src/position.h
#	src/types.h
#	src/ucioption.cpp
#	stockfish.md
 2020-08-08 15:55:42 +09:00
nodchip 1abae04ceb Fixed Makefile. 2020-08-07 23:00:11 +09:00
nodchip bf7d02578e Fixed build errors. 2020-08-07 22:47:45 +09:00
nodchip 1c23465383 Moved the nnue folder. 2020-08-07 22:34:53 +09:00
nodchipandJoost VandeVondele 84f3e86790 Add NNUE evaluation 
This patch ports the efficiently updatable neural network (NNUE) evaluation to Stockfish.

Both the NNUE and the classical evaluations are available, and can be used to
assign a value to a position that is later used in alpha-beta (PVS) search to find the
best move. The classical evaluation computes this value as a function of various chess
concepts, handcrafted by experts, tested and tuned using fishtest. The NNUE evaluation
computes this value with a neural network based on basic inputs. The network is optimized
and trained on the evalutions of millions of positions at moderate search depth.

The NNUE evaluation was first introduced in shogi, and ported to Stockfish afterward.
It can be evaluated efficiently on CPUs, and exploits the fact that only parts
of the neural network need to be updated after a typical chess move.
[The nodchip repository](https://github.com/nodchip/Stockfish) provides additional
tools to train and develop the NNUE networks.

This patch is the result of contributions of various authors, from various communities,
including: nodchip, ynasu87, yaneurao (initial port and NNUE authors), domschl, FireFather,
rqs, xXH4CKST3RXx, tttak, zz4032, joergoster, mstembera, nguyenpham, erbsenzaehler,
dorzechowski, and vondele.

This new evaluation needed various changes to fishtest and the corresponding infrastructure,
for which tomtor, ppigazzini, noobpwnftw, daylen, and vondele are gratefully acknowledged.

The first networks have been provided by gekkehenker and sergiovieri, with the latter
net (nn-97f742aaefcd.nnue) being the current default.

The evaluation function can be selected at run time with the `Use NNUE` (true/false) UCI option,
provided the `EvalFile` option points the the network file (depending on the GUI, with full path).

The performance of the NNUE evaluation relative to the classical evaluation depends somewhat on
the hardware, and is expected to improve quickly, but is currently on > 80 Elo on fishtest:

60000 @ 10+0.1 th 1
https://tests.stockfishchess.org/tests/view/5f28fe6ea5abc164f05e4c4c
ELO: 92.77 +-2.1 (95%) LOS: 100.0%
Total: 60000 W: 24193 L: 8543 D: 27264
Ptnml(0-2): 609, 3850, 9708, 10948, 4885

40000 @ 20+0.2 th 8
https://tests.stockfishchess.org/tests/view/5f290229a5abc164f05e4c58
ELO: 89.47 +-2.0 (95%) LOS: 100.0%
Total: 40000 W: 12756 L: 2677 D: 24567
Ptnml(0-2): 74, 1583, 8550, 7776, 2017

At the same time, the impact on the classical evaluation remains minimal, causing no significant
regression:

sprt @ 10+0.1 th 1
https://tests.stockfishchess.org/tests/view/5f2906a2a5abc164f05e4c5b
LLR: 2.94 (-2.94,2.94) {-6.00,-4.00}
Total: 34936 W: 6502 L: 6825 D: 21609
Ptnml(0-2): 571, 4082, 8434, 3861, 520

sprt @ 60+0.6 th 1
https://tests.stockfishchess.org/tests/view/5f2906cfa5abc164f05e4c5d
LLR: 2.93 (-2.94,2.94) {-6.00,-4.00}
Total: 10088 W: 1232 L: 1265 D: 7591
Ptnml(0-2): 49, 914, 3170, 843, 68

The needed networks can be found at https://tests.stockfishchess.org/nns
It is recommended to use the default one as indicated by the `EvalFile` UCI option.

Guidelines for testing new nets can be found at
https://github.com/glinscott/fishtest/wiki/Creating-my-first-test#nnue-net-tests

Integration has been discussed in various issues:
https://github.com/official-stockfish/Stockfish/issues/2823
https://github.com/official-stockfish/Stockfish/issues/2728

The integration branch will be closed after the merge:
https://github.com/official-stockfish/Stockfish/pull/2825
https://github.com/official-stockfish/Stockfish/tree/nnue-player-wip

closes https://github.com/official-stockfish/Stockfish/pull/2912

This will be an exciting time for computer chess, looking forward to seeing the evolution of
this approach.

Bench: 4746616
 2020-08-06 16:37:45 +02:00
nodchip c0e1235fef Added a description to Makefile. 2020-07-20 17:36:09 +09:00
nodchip fbdb373b64 Changed to set the binary directory to the current working directory. 2020-07-20 17:17:50 +09:00
nodchip afd7d0ea4d Fixed a bug that Makefile specifies -mpopcnt for armv8-a. 2020-07-19 18:34:35 +09:00
nodchip 36092b855a Removed the x86-64-ssse3-popcnt architecture. 2020-07-19 14:17:35 +09:00
nodchip 3bbe4802b1 Removed the sse41-popcnt architecture. 2020-07-19 14:02:49 +09:00
nodchip c001a4e62d Revert "Removed x86-64-ssse3-popcnt and x86-64-sse41-popcnt." 
This reverts commit 92c2167481.
 2020-07-19 13:58:19 +09:00
nodchip 92c2167481 Removed x86-64-ssse3-popcnt and x86-64-sse41-popcnt. 2020-07-19 12:52:20 +09:00
nodchip a4786db4c2 Added support for architectures which supports SSE3+POPCNT, SSSE3+POPCNT and SSE41+POPCNT. 2020-07-19 12:41:50 +09:00
nodchip c24ad8d8b5 Supported sse3 build. 2020-07-19 12:26:37 +09:00
nodchip 7a13d4ed60 Changed the default eval file path so that more GUIs can use Stockfish NNUE. 2020-07-17 15:40:01 +09:00
nodchipandGitHub 2b821682aa Update README.md 2020-07-17 11:55:30 +09:00
nodchip 7f4b72cdfd Merge branch 'master' of github.com:nodchip/Stockfish 2020-07-13 22:25:56 +09:00
nodchip 686a5a0df9 Fixed a bug that gensfen command does not accept the use_draw_in_training_data_generation option. 2020-07-13 22:25:23 +09:00
nodchipandGitHub ae4db5ebfd Merge pull request #45 from joergoster/sf-nnue-update 
Sf nnue update
 2020-07-11 19:17:38 +09:00
nodchip d7c358cf19 Fixed descriptions and sanity checks in Makefile. 2020-07-10 16:55:32 +09:00
nodchip bc6a8d09e9 Unified the nnue-learn and nnue-learn-use-blas targets into nnue-learn. 2020-07-10 16:17:35 +09:00
nodchip df05ecb1d5 Added halfkp_384x2-32-32. 2020-07-10 16:14:19 +09:00
nodchip 1de1eb2d0d Refactoring: Restructured the architecture list in Makefile. 2020-07-10 16:13:21 +09:00
nodchipandGitHub 272f0f88c3 Merge pull request #40 from FireFather/master 
Update evaluate_nnue_learner.cpp
 2020-07-08 20:52:39 +09:00
nodchip 288fdc5597 Added "nodes" option to the "gensfen" command to specify the number of the nodes to be searched. 2020-07-06 17:38:43 +09:00
nodchip 85c802d0b9 Revert "use winning_percentage_wdl in learn" 
This reverts commit c964e902c5.

# Conflicts:
#	src/uci.cpp
 2020-07-06 11:07:46 +09:00
nodchipandGitHub 3b535b5ade Merge pull request #36 from tttak/WDL_20200703b 
use winning_percentage_wdl in learn
 2020-07-04 07:36:33 +09:00
nodchipandGitHub cea5240909 Merge branch 'master' into WDL_20200703b 2020-07-04 07:36:16 +09:00
nodchip c8262f8aec Fixed a compile error. 2020-06-30 15:58:51 +09:00
nodchipandGitHub 9dc62809c8 Merge pull request #24 from FireFather/master 
Comment translation
 2020-06-28 11:43:53 +09:00
nodchipandGitHub 123dd68452 Merge branch 'master' into master 2020-06-28 11:42:23 +09:00
nodchip f1a8580118 Merge branch 'master' of github.com:nodchip/Stockfish 2020-06-28 11:37:42 +09:00
nodchip 87c8b324f8 Simplified source code to estimate the winning ratio from an eval value. 
We need to adjust the eta again after this commit is pushed.
 2020-06-28 11:37:15 +09:00
nodchipandGitHub 6324c2de75 Merge pull request #23 from joergoster/linux_sse41 
Linux sse41
 2020-06-28 10:29:14 +09:00
nodchip cb8accada2 Merge branch 'master' of github.com:nodchip/Stockfish 2020-06-27 22:19:33 +09:00
nodchip 13eb540020 Changed the formula to calculate winning ratio to 1/(1+10^(-Eval/4)). 2020-06-27 22:19:22 +09:00
nodchip aa2dc962f5 Added use_draw_in_training_data_generation option to write out draw games to the training data. 2020-06-27 14:00:12 +09:00
nodchipandGitHub 1c8a931309 Merge pull request #18 from joergoster/sf-nnue-nodchip 
Update to SF master
 2020-06-25 23:09:45 +09:00
nodchipandGitHub ff31d92b94 Merge pull request #14 from joergoster/sf-nnue-nodchip 
Update to SF master
 2020-06-25 10:38:16 +09:00
nodchipandGitHub 7818d23afb Merge pull request #13 from zz4032/linux-identifier 
Linux identifier corrected.
 2020-06-24 09:35:44 +09:00
nodchip 0abd692543 Fixed a build error on Visual Studio. 2020-06-24 09:33:46 +09:00
nodchip 999f5ec446 COMP=msys2を指定できるようにした 2020-06-24 00:47:34 +09:00
nodchip 43e78187d7 ARCH=x86-64-avx2を指定できるようにした 2020-06-24 00:27:45 +09:00
nodchip 76b0de40a1 コンパイラーオプションを-std=c++14から-std=c++17に変更した 2020-06-23 23:47:59 +09:00
nodchip c7884470fb Merge branch 'master' of github.com:nodchip/Stockfish 2020-06-22 10:28:02 +09:00
nodchip 6c7a594362 Added "-Wl,-s" option. 2020-06-22 10:27:52 +09:00
nodchip 33772a0418 コンパイルエラーを修正した 2020-06-08 23:46:06 +09:00
nodchip 5c936572e9 Merge branch 'master' of github.com:official-stockfish/Stockfish 
# Conflicts:
#	src/Makefile
#	src/position.cpp
#	src/position.h
#	src/search.cpp
#	src/types.h
#	src/uci.cpp
 2020-06-08 23:09:51 +09:00
nodchip d23f96d156 No pruning at low plies. 
This makes those very early depths a bit more reliable.

Thanks, joergoster!

https://github.com/joergoster/Stockfish-NNUE/commit/be7f37187b85b8093ae0741909cbfd7b2bc76871
 2020-06-06 18:50:15 +09:00
nodchip 2523f72ff9 盤上から取り除かれた駒に関する差分計算を省き、高速化した 2020-06-03 23:32:08 +09:00
nodchip a85e3055f4 いくつかのターゲットでOpenMPを有効にした 2020-06-02 00:13:35 +09:00
nodchip 6703ec8ab0 nnue-gen-sfen-from-original-evalターゲットを追加した 2020-05-30 09:50:29 +09:00
nodchip dd9818c2c1 Added "-static" compiler option. 2020-05-29 09:36:24 +09:00
nodchip f18acf97ed Added the "nnue" target. 
Fixed build errors on the "nnue-learn-use-blas" target.
 2020-05-28 10:08:51 +09:00
Hisayori Noda e2165155d1 Enabled halfkp_256x2-32-32.h. 2019-07-11 22:47:55 +09:00
Hisayori Noda 1d5f79db1c Merge branch 'master' of github.com:nodchip/Stockfish into k-p_256x2-32-32 2019-07-10 08:18:57 +09:00
Hisayori Noda 2d70487caa Enabled k-p_256x2-32-32. 2019-07-08 19:02:09 +09:00
Hisayori Noda b300a9d43e Enabled halfkp_256x2-32-32. 2019-07-07 21:44:02 +09:00
Hisayori Noda 747d98bf1b Added halfkp-cr-ep_256x2-32-32 architecture. 2019-07-07 20:31:54 +09:00
Hisayori Noda df827ea7ee Added enpassant feature. Added k-p-cr-ep_256x2-32-32 architecture. 2019-07-07 19:24:46 +09:00
Hisayori Noda 92052bc16b Fixed build errors. 2019-07-07 17:22:07 +09:00
Hisayori Noda 5c0037de7f Added the castling right feature. Added k-p-cr_256x2-32-32 architecture. 2019-07-07 17:02:34 +09:00
Hisayori Noda 09e529edd3 Added a hack to avoid crash with binaries compiled by g++ on MSYS2. 2019-07-05 00:22:41 +09:00
Hisayori Noda 89e846c476 Fixed a bug that Learner::qsearch() recognizes stalemate as checkmated. 2019-07-04 23:44:58 +09:00
Hisayori Noda 81262320c3 Revert "Changed the constant value to calculate the winning percentage." 
This reverts commit 00f84ed99a.
 2019-06-30 11:29:43 +09:00
Hisayori Noda 00f84ed99a Changed the constant value to calculate the winning percentage. 2019-06-26 08:48:48 +09:00
Hisayori Noda 26271586cb Added #if to fix compile errors. 2019-06-24 23:18:17 +09:00
Hisayori Noda a413bf7aad Added hack to avoid crash during machine learning. 2019-06-24 23:17:46 +09:00
Hisayori Noda 9a73df7379 Added test commands. 2019-06-22 00:40:46 +09:00
Hisayori Noda 5772509e8b Disabled TT when EVAL_LEARN is enabled. 2019-06-22 00:40:25 +09:00
Hisayori Noda 641724e3a5 Added debug code. 2019-06-22 00:39:21 +09:00
Hisayori Noda 57ead90f18 Fixed a bug that the game play is invalid. 2019-06-22 00:38:24 +09:00
Hisayori Noda 07dc336b0f Added validation logic. 2019-06-22 00:37:59 +09:00
Hisayori Noda 84a96a3d9c Fixed a compilation error. 2019-06-22 00:37:31 +09:00
Hisayori Noda 998d8721bd Fixed a bug that White and Black are reversed. 2019-06-22 00:37:10 +09:00
Hisayori Noda 9dab4660ce Added source files. 2019-06-22 00:36:42 +09:00
Hisayori Noda 90ef97dcbd Fixed crash bugs. 2019-06-20 00:25:40 +09:00
Hisayori Noda 24576d77ab Merged uci parse logic. 2019-06-18 21:19:51 +09:00
Hisayori Noda f58d616198 Fixed compile errors when EVAL_LEARN or EVAL_NNUE are defined. 2019-06-18 20:28:50 +09:00
Hisayori Noda bcd6985871 Merged the training data generator and the machine learning logic from YaneuraOu. 2019-06-18 08:48:05 +09:00
Hisayori Noda 87445881ec Added #ifdef statements to switch the legacy evaluation function and NNUE evaluation function. 2019-06-16 11:11:16 +09:00
Hisayori Noda 48bfe86d27 Implemented the logic to update Eval List and Dirty Pieces. 2019-06-16 10:33:53 +09:00
Hisayori Noda b330602cdc Fixed compile errors. 2019-06-15 17:08:47 +09:00
Hisayori Noda 9964fbbe25 Reverted evaluate.cpp. 2019-06-15 11:46:54 +09:00
Hisayori Noda db02ddcc90 Added files for NNUE. 2019-06-09 10:40:12 +09:00