Latest abrok Development Version August 16th New NNUE architecture and net (SF The gift that keeps on giving )

Windows x64 for Haswell CPUs
Windows x64 for modern computers + AVX2
Windows x64 for modern computers
Windows x64 + SSSE3
Windows x64
Windows 32
Linux x64 for Haswell CPUs
Linux x64 for modern computers + AVX2
Linux x64 for modern computers
Linux x64 + SSSE3
Linux x64

Author: Tomasz Sobczyk
Date: Sun Aug 15 12:05:43 2021 +0200
Timestamp: 1629021943

New NNUE architecture and net

Introduces a new NNUE network architecture and associated network parameters

The summary of the changes:

* Position for each perspective mirrored such that the king is on e..h files. Cuts the feature transformer size in half, while preserving enough knowledge to be good. See https://docs.google.com/document/d/1gTlrr02qSNKiXNZ_SuO4-RjK4MXBiFlLE6jvNqqMkAY/edit#heading=h.b40q4rb1w7on.
* The number of neurons after the feature transformer increased two-fold, to 1024x2. This is possibly mostly due to the now very optimized feature transformer update code.
* The number of neurons after the second layer is reduced from 16 to 8, to reduce the speed impact. This, perhaps surprisingly, doesn't harm the strength much. See https://docs.google.com/document/d/1gTlrr02qSNKiXNZ_SuO4-RjK4MXBiFlLE6jvNqqMkAY/edit#heading=h.6qkocr97fezq

The AffineTransform code did not work out-of-the box with the smaller number of neurons after the second layer, so some temporary changes have been made to add a special case for InputDimensions == 8. Also additional 0 padding is added to the output for some archs that cannot process inputs by <=8 (SSE2, NEON). VNNI uses an implementation that can keep all outputs in the registers while reducing the number of loads by 3 for each 16 inputs, thanks to the reduced number of output neurons. However GCC is particularily bad at optimization here (and perhaps why the current way the affine transform is done even passed sprt) (see https://docs.google.com/document/d/1gTlrr02qSNKiXNZ_SuO4-RjK4MXBiFlLE6jvNqqMkAY/edit# for details) and more work will be done on this in the following days. I expect the current VNNI implementation to be improved and extended to other architectures.

The network was trained with a slightly modified version of the pytorch trainer (https://github.com/glinscott/nnue-pytorch); the changes are in https://github.com/glinscott/nnue-pytorch/pull/143

The training utilized 2 datasets.

dataset A - https://drive.google.com/file/d/1VlhnHL8f-20AXhGkILujnNXHwy9T-MQw/view?usp=sharing
dataset B - as described in https://github.com/official-stockfish/Stockfish/commit/ba01f4b95448bcb324755f4dd2a632a57c6e67bc

The training process was as following:

train on dataset A for 350 epochs, take the best net in terms of elo at 20k nodes per move (it's fine to take anything from later stages of training).
convert the .ckpt to .pt
--resume-from-model from the .pt file, train on dataset B for <600 epochs, take the best net. Lambda=0.8, applied before the loss function.

The first training command:

python3 train.py \
../nnue-pytorch-training/data/large_gensfen_multipvdiff_100_d9.binpack \
../nnue-pytorch-training/data/large_gensfen_multipvdiff_100_d9.binpack \
--gpus "$3," \
--threads 1 \
--num-workers 1 \
--batch-size 16384 \
--progress_bar_refresh_rate 20 \
--smart-fen-skipping \
--random-fen-skipping 3 \
--features=HalfKAv2_hm^ \
--lambda=1.0 \
--max_epochs=600 \
--default_root_dir ../nnue-pytorch-training/experiment_$1/run_$2

The second training command:

python3 serialize.py \
--features=HalfKAv2_hm^ \
../nnue-pytorch-training/experiment_131/run_6/default/version_0/checkpoints/epoch-499.ckpt \
../nnue-pytorch-training/experiment_$1/base/base.pt

python3 train.py \
../nnue-pytorch-training/data/michael_commit_b94a65.binpack \
../nnue-pytorch-training/data/michael_commit_b94a65.binpack \
--gpus "$3," \
--threads 1 \
--num-workers 1 \
--batch-size 16384 \
--progress_bar_refresh_rate 20 \
--smart-fen-skipping \
--random-fen-skipping 3 \
--features=HalfKAv2_hm^ \
--lambda=0.8 \
--max_epochs=600 \
--resume-from-model ../nnue-pytorch-training/experiment_$1/base/base.pt \
--default_root_dir ../nnue-pytorch-training/experiment_$1/run_$2

STC: https://tests.stockfishchess.org/tests/view/611120b32a8a49ac5be798c4

LLR: 2.97 (-2.94,2.94) <-0.50,2.50>
Total: 22480 W: 2434 L: 2251 D: 17795 Elo +2.83
Ptnml(0-2): 101, 1736, 7410, 1865, 128

LTC: https://tests.stockfishchess.org/tests/view/611152b32a8a49ac5be798ea

LLR: 2.93 (-2.94,2.94) <0.50,3.50>
Total: 9776 W: 442 L: 333 D: 9001 Elo +3.87
Ptnml(0-2): 5, 295, 4180, 402, 6

closes https://github.com/official-stockfish/Stockfish/pull/3646

bench: 5189338
see source

_________________
looking for Bettina

» Latest abrok Development Version May 18th (SF The gift that keeps on giving ) New NNUE architecture and net
» Latest abrok Development Version May 14th * Including Update NNUE architecture to SFNNv5 * (SF The gift that keeps on giving
» Latest abrok Development Version August 5th Update default net to nn-46832cfbead3.nnue (SF The gift that keeps on giving )
» Latest abrok Development Version July 1st Update NNUE architecture to SFNNv7 with larger L1 size of 2048 (SF The gift that keeps on giving
» Latest abrok Development Version August 18th :!!: Update default net to nn-ac5605a608d6.nnue (SF The gift that keeps on giving )

Latest abrok Development Version August 16th ** New NNUE architecture and net **(SF The gift that keeps on giving )

Latest abrok Development Version August 16th New NNUE architecture and net (SF The gift that keeps on giving )