mirror of
https://github.com/sockspls/badfish
synced 2025-04-29 16:23:09 +00:00

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
155 lines
5.2 KiB
C++
155 lines
5.2 KiB
C++
/*
|
|
Stockfish, a UCI chess playing engine derived from Glaurung 2.1
|
|
Copyright (C) 2004-2020 The Stockfish developers (see AUTHORS file)
|
|
|
|
Stockfish is free software: you can redistribute it and/or modify
|
|
it under the terms of the GNU General Public License as published by
|
|
the Free Software Foundation, either version 3 of the License, or
|
|
(at your option) any later version.
|
|
|
|
Stockfish is distributed in the hope that it will be useful,
|
|
but WITHOUT ANY WARRANTY; without even the implied warranty of
|
|
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
|
GNU General Public License for more details.
|
|
|
|
You should have received a copy of the GNU General Public License
|
|
along with this program. If not, see <http://www.gnu.org/licenses/>.
|
|
*/
|
|
|
|
#include <cstring> // For std::memset
|
|
#include <iostream>
|
|
#include <thread>
|
|
|
|
#include "bitboard.h"
|
|
#include "misc.h"
|
|
#include "thread.h"
|
|
#include "tt.h"
|
|
#include "uci.h"
|
|
|
|
TranspositionTable TT; // Our global transposition table
|
|
|
|
/// TTEntry::save() populates the TTEntry with a new node's data, possibly
|
|
/// overwriting an old position. Update is not atomic and can be racy.
|
|
|
|
void TTEntry::save(Key k, Value v, bool pv, Bound b, Depth d, Move m, Value ev) {
|
|
|
|
// Preserve any existing move for the same position
|
|
if (m || (uint16_t)k != key16)
|
|
move16 = (uint16_t)m;
|
|
|
|
// Overwrite less valuable entries
|
|
if ((uint16_t)k != key16
|
|
|| d - DEPTH_OFFSET > depth8 - 4
|
|
|| b == BOUND_EXACT)
|
|
{
|
|
assert(d >= DEPTH_OFFSET);
|
|
|
|
key16 = (uint16_t)k;
|
|
value16 = (int16_t)v;
|
|
eval16 = (int16_t)ev;
|
|
genBound8 = (uint8_t)(TT.generation8 | uint8_t(pv) << 2 | b);
|
|
depth8 = (uint8_t)(d - DEPTH_OFFSET);
|
|
}
|
|
}
|
|
|
|
|
|
/// TranspositionTable::resize() sets the size of the transposition table,
|
|
/// measured in megabytes. Transposition table consists of a power of 2 number
|
|
/// of clusters and each cluster consists of ClusterSize number of TTEntry.
|
|
|
|
void TranspositionTable::resize(size_t mbSize) {
|
|
|
|
Threads.main()->wait_for_search_finished();
|
|
|
|
aligned_ttmem_free(mem);
|
|
|
|
clusterCount = mbSize * 1024 * 1024 / sizeof(Cluster);
|
|
table = static_cast<Cluster*>(aligned_ttmem_alloc(clusterCount * sizeof(Cluster), mem));
|
|
if (!mem)
|
|
{
|
|
std::cerr << "Failed to allocate " << mbSize
|
|
<< "MB for transposition table." << std::endl;
|
|
exit(EXIT_FAILURE);
|
|
}
|
|
|
|
clear();
|
|
}
|
|
|
|
|
|
/// TranspositionTable::clear() initializes the entire transposition table to zero,
|
|
// in a multi-threaded way.
|
|
|
|
void TranspositionTable::clear() {
|
|
|
|
std::vector<std::thread> threads;
|
|
|
|
for (size_t idx = 0; idx < Options["Threads"]; ++idx)
|
|
{
|
|
threads.emplace_back([this, idx]() {
|
|
|
|
// Thread binding gives faster search on systems with a first-touch policy
|
|
if (Options["Threads"] > 8)
|
|
WinProcGroup::bindThisThread(idx);
|
|
|
|
// Each thread will zero its part of the hash table
|
|
const size_t stride = size_t(clusterCount / Options["Threads"]),
|
|
start = size_t(stride * idx),
|
|
len = idx != Options["Threads"] - 1 ?
|
|
stride : clusterCount - start;
|
|
|
|
std::memset(&table[start], 0, len * sizeof(Cluster));
|
|
});
|
|
}
|
|
|
|
for (std::thread& th : threads)
|
|
th.join();
|
|
}
|
|
|
|
|
|
/// TranspositionTable::probe() looks up the current position in the transposition
|
|
/// table. It returns true and a pointer to the TTEntry if the position is found.
|
|
/// Otherwise, it returns false and a pointer to an empty or least valuable TTEntry
|
|
/// to be replaced later. The replace value of an entry is calculated as its depth
|
|
/// minus 8 times its relative age. TTEntry t1 is considered more valuable than
|
|
/// TTEntry t2 if its replace value is greater than that of t2.
|
|
|
|
TTEntry* TranspositionTable::probe(const Key key, bool& found) const {
|
|
|
|
TTEntry* const tte = first_entry(key);
|
|
const uint16_t key16 = (uint16_t)key; // Use the low 16 bits as key inside the cluster
|
|
|
|
for (int i = 0; i < ClusterSize; ++i)
|
|
if (!tte[i].key16 || tte[i].key16 == key16)
|
|
{
|
|
tte[i].genBound8 = uint8_t(generation8 | (tte[i].genBound8 & 0x7)); // Refresh
|
|
|
|
return found = (bool)tte[i].key16, &tte[i];
|
|
}
|
|
|
|
// Find an entry to be replaced according to the replacement strategy
|
|
TTEntry* replace = tte;
|
|
for (int i = 1; i < ClusterSize; ++i)
|
|
// Due to our packed storage format for generation and its cyclic
|
|
// nature we add 263 (256 is the modulus plus 7 to keep the unrelated
|
|
// lowest three bits from affecting the result) to calculate the entry
|
|
// age correctly even after generation8 overflows into the next cycle.
|
|
if ( replace->depth8 - ((263 + generation8 - replace->genBound8) & 0xF8)
|
|
> tte[i].depth8 - ((263 + generation8 - tte[i].genBound8) & 0xF8))
|
|
replace = &tte[i];
|
|
|
|
return found = false, replace;
|
|
}
|
|
|
|
|
|
/// TranspositionTable::hashfull() returns an approximation of the hashtable
|
|
/// occupation during a search. The hash is x permill full, as per UCI protocol.
|
|
|
|
int TranspositionTable::hashfull() const {
|
|
|
|
int cnt = 0;
|
|
for (int i = 0; i < 1000; ++i)
|
|
for (int j = 0; j < ClusterSize; ++j)
|
|
cnt += (table[i].entry[j].genBound8 & 0xF8) == generation8;
|
|
|
|
return cnt / ClusterSize;
|
|
}
|