kate/BadFish
1
0
Fork 0
mirror of https://github.com/sockspls/badfish synced 2025-04-30 08:43:09 +00:00
Code Issues Releases Wiki Activity

Small cleanups (2)

Browse source 
- fix a small compile error under MSVC
- improve sigmoid comment and assert
- fix formatting in README.md

closes https://github.com/official-stockfish/Stockfish/pull/3960

No functional change
This commit is contained in:
mstembera 2022-03-12 07:00:58 -08:00 • committed by Stéphane Nicolet
parent 004ea2c25e
commit f3a2296e59
3 changed files with 44 additions and 36 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

71
README.md
View file

@ -10,23 +10,28 @@ Cute Chess, eboard, Arena, Sigma Chess, Shredder, Chess Partner or Fritz) in ord
to be used comfortably. Read the documentation for your GUI of choice for information to be used comfortably. Read the documentation for your GUI of choice for information
about how to use Stockfish with it. about how to use Stockfish with it.
The Stockfish engine features two evaluation functions for chess. The Stockfish engine features two evaluation functions for chess. The efficiently
The efficiently updatable neural network (NNUE) based evaluation is the default and by far the strongest. updatable neural network (NNUE) based evaluation is the default and by far the strongest.
The classical evaluation based on handcrafted terms remains available. The classical evaluation based on handcrafted terms remains available. The strongest
The strongest network is integrated in the binary and downloaded automatically during the build process. network is integrated in the binary and downloaded automatically during the build process.
The NNUE evaluation benefits from the vector intrinsics available on most CPUs (sse2, avx2, neon, or similar). The NNUE evaluation benefits from the vector intrinsics available on most CPUs (sse2,
avx2, neon, or similar).
## Files ## Files
This distribution of Stockfish consists of the following files: This distribution of Stockfish consists of the following files:
* [Readme.md](https://github.com/official-stockfish/Stockfish/blob/master/README.md), the file you are currently reading. * [Readme.md](https://github.com/official-stockfish/Stockfish/blob/master/README.md),
the file you are currently reading.
* [Copying.txt](https://github.com/official-stockfish/Stockfish/blob/master/Copying.txt), a text file containing the GNU General Public License version 3. * [Copying.txt](https://github.com/official-stockfish/Stockfish/blob/master/Copying.txt),
a text file containing the GNU General Public License version 3.
* [AUTHORS](https://github.com/official-stockfish/Stockfish/blob/master/AUTHORS), a text file with the list of authors for the project * [AUTHORS](https://github.com/official-stockfish/Stockfish/blob/master/AUTHORS),
a text file with the list of authors for the project
* [src](https://github.com/official-stockfish/Stockfish/tree/master/src), a subdirectory containing the full source code, including a Makefile * [src](https://github.com/official-stockfish/Stockfish/tree/master/src),
a subdirectory containing the full source code, including a Makefile
that can be used to compile Stockfish on Unix-like systems. that can be used to compile Stockfish on Unix-like systems.
* a file with the .nnue extension, storing the neural network for the NNUE * a file with the .nnue extension, storing the neural network for the NNUE
@ -67,9 +72,9 @@ change them via a chess GUI. This is a list of available UCI options in Stockfis
* #### EvalFile * #### EvalFile
The name of the file of the NNUE evaluation parameters. Depending on the GUI the The name of the file of the NNUE evaluation parameters. Depending on the GUI the
filename might have to include the full path to the folder/directory that contains the file. filename might have to include the full path to the folder/directory that contains
Other locations, such as the directory that contains the binary and the working directory, the file. Other locations, such as the directory that contains the binary and the
are also searched. working directory, are also searched.
* #### UCI_AnalyseMode * #### UCI_AnalyseMode
An option handled by your GUI. An option handled by your GUI.
@ -137,8 +142,9 @@ change them via a chess GUI. This is a list of available UCI options in Stockfis
For developers the following non-standard commands might be of interest, mainly useful for debugging: For developers the following non-standard commands might be of interest, mainly useful for debugging:
* #### bench *ttSize threads limit fenFile limitType evalType* * #### bench *ttSize threads limit fenFile limitType evalType*
Performs a standard benchmark using various options. The signature of a version (standard node Performs a standard benchmark using various options. The signature of a version
count) is obtained using all defaults. `bench` is currently `bench 16 1 13 default depth mixed`. (standard node count) is obtained using all defaults. `bench` is currently
`bench 16 1 13 default depth mixed`.
* #### compiler * #### compiler
Give information about the compiler and environment used for building a binary. Give information about the compiler and environment used for building a binary.
@ -174,26 +180,27 @@ on the evaluations of millions of positions at moderate search depth.
The NNUE evaluation was first introduced in shogi, and ported to Stockfish afterward. 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 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. of the neural network need to be updated after a typical chess move.
[The nodchip repository](https://github.com/nodchip/Stockfish) provided the first version of [The nodchip repository](https://github.com/nodchip/Stockfish) provided the first
the needed tools to train and develop the NNUE networks. Today, more advanced training tools are available version of the needed tools to train and develop the NNUE networks. Today, more
in [the nnue-pytorch repository](https://github.com/glinscott/nnue-pytorch/), while data generation tools advanced training tools are available in
are available in [a dedicated branch](https://github.com/official-stockfish/Stockfish/tree/tools). [the nnue-pytorch repository](https://github.com/glinscott/nnue-pytorch/),
while data generation tools are available in
[a dedicated branch](https://github.com/official-stockfish/Stockfish/tree/tools).
On CPUs supporting modern vector instructions On CPUs supporting modern vector instructions (avx2 and similar), the NNUE evaluation
(avx2 and similar), the NNUE evaluation results in much stronger playing strength, even results in much stronger playing strength, even if the nodes per second computed by
if the nodes per second computed by the engine is somewhat lower (roughly 80% of nps the engine is somewhat lower (roughly 80% of nps is typical).
is typical).
Notes: Notes:
1) the NNUE evaluation depends on the Stockfish binary and the network parameter 1) the NNUE evaluation depends on the Stockfish binary and the network parameter file
file (see the EvalFile UCI option). Not every parameter file is compatible with a given (see the EvalFile UCI option). Not every parameter file is compatible with a given
Stockfish binary, but the default value of the EvalFile UCI option is the name of a network Stockfish binary, but the default value of the EvalFile UCI option is the name of a
that is guaranteed to be compatible with that binary. network that is guaranteed to be compatible with that binary.
2) to use the NNUE evaluation, the additional data file with neural network parameters 2) to use the NNUE evaluation, the additional data file with neural network parameters
needs to be available. Normally, this file is already embedded in the binary or it needs to be available. Normally, this file is already embedded in the binary or it can
can be downloaded. The filename for the default (recommended) net can be found as the default be downloaded. The filename for the default (recommended) net can be found as the default
value of the `EvalFile` UCI option, with the format `nn-[SHA256 first 12 digits].nnue` value of the `EvalFile` UCI option, with the format `nn-[SHA256 first 12 digits].nnue`
(for instance, `nn-c157e0a5755b.nnue`). This file can be downloaded from (for instance, `nn-c157e0a5755b.nnue`). This file can be downloaded from
``` ```
@ -321,10 +328,10 @@ it (either by itself or as part of some bigger software package), or
using it as the starting point for a software project of your own. using it as the starting point for a software project of your own.
The only real limitation is that whenever you distribute Stockfish in The only real limitation is that whenever you distribute Stockfish in
some way, you MUST always include the license, the full source code, or a pointer some way, you MUST always include the license and the full source code
to where the source code can be found, to generate the exact binary (or a pointer to where the source code can be found) to generate the
you are distributing. If you make any changes to the source code, exact binary you are distributing. If you make any changes to the
these changes must also be made available under the GPL v3. source code, these changes must also be made available under the GPL v3.
For full details, read the copy of the GPL v3 found in the file named For full details, read the copy of the GPL v3 found in the file named
[*Copying.txt*](https://github.com/official-stockfish/Stockfish/blob/master/Copying.txt). [*Copying.txt*](https://github.com/official-stockfish/Stockfish/blob/master/Copying.txt).

5
src/misc.h
View file

@ -152,7 +152,7 @@ private:
/// - the slope can be adjusted using C > 0, smaller C giving a steeper sigmoid /// - the slope can be adjusted using C > 0, smaller C giving a steeper sigmoid
/// - the slope of the sigmoid when t = x0 is P/(Q*C) /// - the slope of the sigmoid when t = x0 is P/(Q*C)
/// - sigmoid is increasing with t when P > 0 and Q > 0 /// - sigmoid is increasing with t when P > 0 and Q > 0
/// - to get a decreasing sigmoid, call with -t, or change sign of P /// - to get a decreasing sigmoid, change sign of P
/// - mean value of the sigmoid is y0 /// - mean value of the sigmoid is y0
/// ///
/// Use <https://www.desmos.com/calculator/jhh83sqq92> to draw the sigmoid /// Use <https://www.desmos.com/calculator/jhh83sqq92> to draw the sigmoid
@ -163,7 +163,8 @@ inline int64_t sigmoid(int64_t t, int64_t x0,
int64_t P, int64_t P,
int64_t Q) int64_t Q)
{ {
assert(C > 0 && Q != 0); assert(C > 0);
assert(Q != 0);
return y0 + P * (t-x0) / (Q * (std::abs(t-x0) + C)) ; return y0 + P * (t-x0) / (Q * (std::abs(t-x0) + C)) ;
} }

4
src/movepick.cpp
View file

@ -107,8 +107,8 @@ void MovePicker::score() {
for (auto& m : *this) for (auto& m : *this)
if constexpr (Type == CAPTURES) if constexpr (Type == CAPTURES)
m.value = 6 * PieceValue[MG][pos.piece_on(to_sq(m))] m.value = 6 * int(PieceValue[MG][pos.piece_on(to_sq(m))])
+ (*captureHistory)[pos.moved_piece(m)][to_sq(m)][type_of(pos.piece_on(to_sq(m)))]; + (*captureHistory)[pos.moved_piece(m)][to_sq(m)][type_of(pos.piece_on(to_sq(m)))];
else if constexpr (Type == QUIETS) else if constexpr (Type == QUIETS)
m.value = (*mainHistory)[pos.side_to_move()][from_to(m)] m.value = (*mainHistory)[pos.side_to_move()][from_to(m)]