Update default net to nn-a3d1bfca1672.nnue

daniel-monroe · vondele · commit ef94f77f8c82 · 2023-07-01T12:59:28.000+02:00
faster permutation of master net weights Activation data taken from https://drive.google.com/drive/folders/1Ec9YuuRx4N03GPnVPoQOW70eucOKngQe?usp=sharing Permutation found using https://github.com/Ergodice/nnue-pytorch/blob/836387a0e5e690431d404158c46648710f13904d/ftperm.py See also official-stockfish/nnue-pytorch#254 The algorithm greedily selects 2- and 3-cycles that can be permuted to increase the number of runs of zeroes. The percent of zero runs from the master net increased from 68.46 to 70.11 from 2-cycles and only increased to 70.32 when considering 3-cycles. Interestingly, allowing both halves of L1 to intermix when creating zero runs can give another 0.5% zero-run density increase with this method. Measured speedup: ``` CPU: 16 x AMD Ryzen 9 3950X 16-Core Processor Result of 50 runs base (./stockfish.master ) = 1561556 +/- 5439 test (./stockfish.patch ) = 1575788 +/- 5427 diff = +14231 +/- 2636 speedup = +0.0091 P(speedup > 0) = 1.0000 ``` closes official-stockfish#4640 No functional change
diff --git a/AUTHORS b/AUTHORS
@@ -48,6 +48,7 @@ clefrks
 Dale Weiler (graphitemaster)
 Daniel Axtens (daxtens)
 Daniel Dugovic (ddugovic)
+Daniel Monroe (Ergodice)
 Dan Schmidt (dfannius)
 Dariusz Orzechowski (dorzechowski)
 David (dav1312)
diff --git a/src/evaluate.h b/src/evaluate.h
@@ -39,7 +39,7 @@ namespace Eval {
   // The default net name MUST follow the format nn-[SHA256 first 12 digits].nnue
   // for the build process (profile-build and fishtest) to work. Do not change the
   // name of the macro, as it is used in the Makefile.
-  #define EvalFileDefaultName   "nn-5af11540bbfe.nnue"
+  #define EvalFileDefaultName   "nn-a3d1bfca1672.nnue"
 
   namespace NNUE {
 
