opelly/Stockfish
1
0
Fork 0
mirror of https://github.com/opelly27/Stockfish.git synced 2026-05-20 07:27:46 +00:00
Code Issues Packages Projects Releases Wiki Activity

Merged the training data generator and the machine learning logic from YaneuraOu.

Browse Source
This commit is contained in:
Hisayori Noda
2019-06-18 08:48:05 +09:00
parent 87445881ec
commit bcd6985871
37 changed files with 6306 additions and 139 deletions
Download Patch File Download Diff File Expand all files Collapse all files
src/eval/nnue/trainer/features/factorizer.h
+6 -8
View File
@@ -3,8 +3,6 @@
#ifndef _NNUE_TRAINER_FEATURES_FACTORIZER_H_
#define _NNUE_TRAINER_FEATURES_FACTORIZER_H_
#include "../../../../config.h"
#if defined(EVAL_NNUE)
#include "../../nnue_common.h"
@@ -29,7 +27,7 @@ class Factorizer {
// 学習用特徴量のインデックスと学習率のスケールを取得する
static void AppendTrainingFeatures(
IndexType base_index, std::vector<TrainingFeature>* training_features) {
ASSERT_LV5(base_index < FeatureType::kDimensions);
assert(base_index < FeatureType::kDimensions);
training_features->emplace_back(base_index);
}
};
@@ -45,8 +43,8 @@ template <typename FeatureType>
IndexType AppendBaseFeature(
FeatureProperties properties, IndexType base_index,
std::vector<TrainingFeature>* training_features) {
ASSERT_LV5(properties.dimensions == FeatureType::kDimensions);
ASSERT_LV5(base_index < FeatureType::kDimensions);
assert(properties.dimensions == FeatureType::kDimensions);
assert(base_index < FeatureType::kDimensions);
training_features->emplace_back(base_index);
return properties.dimensions;
}
@@ -59,14 +57,14 @@ IndexType InheritFeaturesIfRequired(
if (!properties.active) {
return 0;
}
ASSERT_LV5(properties.dimensions == Factorizer<FeatureType>::GetDimensions());
ASSERT_LV5(base_index < FeatureType::kDimensions);
assert(properties.dimensions == Factorizer<FeatureType>::GetDimensions());
assert(base_index < FeatureType::kDimensions);
const auto start = training_features->size();
Factorizer<FeatureType>::AppendTrainingFeatures(
base_index, training_features);
for (auto i = start; i < training_features->size(); ++i) {
auto& feature = (*training_features)[i];
ASSERT_LV5(feature.GetIndex() < Factorizer<FeatureType>::GetDimensions());
assert(feature.GetIndex() < Factorizer<FeatureType>::GetDimensions());
feature.ShiftIndex(index_offset);
}
return properties.dimensions;