Jonathan D.

@Sopel97 Thanks for this! I'll take a closer look later.

@Sopel97 What's the use of KingBuckets[64]? https://github.com/official-stockfish/Stockfish/commit/d61d38586ee35fd4d93445eb547e4af27cc86e6b

@Sopel97 I'm having this error after trying to run `compile_data_loader.bat` (env) (base) C:\Users\User\nnue-pytorch>compile_data_loader.bat (env) (base) C:\Users\User\nnue-pytorch>cmake . -Bbuild -DCMAKE_BUILD_TYPE=RelWithDebInfo -DCMAKE_INSTALL_PREFIX="./" CMake Error at CMakeLists.txt:3 (project): Running 'nmake' '-?' failed with:...

`cmake_minimum_required(VERSION 3.0)` `project(training_data_loader)` I have the training_data_loader file 

@Sopel97 I downloaded the wrongNNUE binpack and it's fine now. There's just some deprecation warning. (env) (base) C:\Users\User\nnue-pytorch>python weight_importance.py --data=wrongNNUE_02_d9.binpack --features=HalfKAv2_hm --pos_n=1024 --best_n=32 --output=out.txt nn-13406b1dcbe0.nnue Done 100 out of 1024...

@Sopel97 What is `{feature_index} ` and `t{output_index}` ? So from the results above, is this the most important weight? `int(Stockfish::Eval::NNUE::featureTransformer->psqtWeights[21062]);` `int psqtW[21062] = {36043};`

@Sopel97 Thanks. Now it's clear to me.

> @SFisGOD On the PSQT note. Have you tried, or considered, training an additive [for example] pawn psqt terms, starting from 0? I have not tried something like that.

@vondele It would be nice if someone implements an auto-LTC for passed STC tests.

I'll gladly take the role. Thanks :)