#ifndef NNUE_ARCHITECTURE_H_INCLUDED
#define NNUE_ARCHITECTURE_H_INCLUDED
-#include <memory>
-
-#include "nnue_common.h"
+#include <cstdint>
+#include <cstring>
+#include <iosfwd>
#include "features/half_ka_v2_hm.h"
-
-#include "layers/affine_transform_sparse_input.h"
#include "layers/affine_transform.h"
+#include "layers/affine_transform_sparse_input.h"
#include "layers/clipped_relu.h"
#include "layers/sqr_clipped_relu.h"
-
-#include "../misc.h"
+#include "nnue_common.h"
namespace Stockfish::Eval::NNUE {
using FeatureSet = Features::HalfKAv2_hm;
// Number of input feature dimensions after conversion
-constexpr IndexType TransformedFeatureDimensions = 2048;
-constexpr IndexType PSQTBuckets = 8;
-constexpr IndexType LayerStacks = 8;
-
-struct Network
-{
- static constexpr int FC_0_OUTPUTS = 15;
- static constexpr int FC_1_OUTPUTS = 32;
-
- Layers::AffineTransformSparseInput<TransformedFeatureDimensions, FC_0_OUTPUTS + 1> fc_0;
- Layers::SqrClippedReLU<FC_0_OUTPUTS + 1> ac_sqr_0;
- Layers::ClippedReLU<FC_0_OUTPUTS + 1> ac_0;
- Layers::AffineTransform<FC_0_OUTPUTS * 2, FC_1_OUTPUTS> fc_1;
- Layers::ClippedReLU<FC_1_OUTPUTS> ac_1;
- Layers::AffineTransform<FC_1_OUTPUTS, 1> fc_2;
-
- // Hash value embedded in the evaluation file
- static constexpr std::uint32_t get_hash_value() {
- // input slice hash
- std::uint32_t hashValue = 0xEC42E90Du;
- hashValue ^= TransformedFeatureDimensions * 2;
-
- hashValue = decltype(fc_0)::get_hash_value(hashValue);
- hashValue = decltype(ac_0)::get_hash_value(hashValue);
- hashValue = decltype(fc_1)::get_hash_value(hashValue);
- hashValue = decltype(ac_1)::get_hash_value(hashValue);
- hashValue = decltype(fc_2)::get_hash_value(hashValue);
-
- return hashValue;
- }
-
- // Read network parameters
- bool read_parameters(std::istream& stream) {
- return fc_0.read_parameters(stream)
- && ac_0.read_parameters(stream)
- && fc_1.read_parameters(stream)
- && ac_1.read_parameters(stream)
- && fc_2.read_parameters(stream);
- }
-
- // Write network parameters
- bool write_parameters(std::ostream& stream) const {
- return fc_0.write_parameters(stream)
- && ac_0.write_parameters(stream)
- && fc_1.write_parameters(stream)
- && ac_1.write_parameters(stream)
- && fc_2.write_parameters(stream);
- }
-
- std::int32_t propagate(const TransformedFeatureType* transformedFeatures)
- {
- struct alignas(CacheLineSize) Buffer
- {
- alignas(CacheLineSize) decltype(fc_0)::OutputBuffer fc_0_out;
- alignas(CacheLineSize) decltype(ac_sqr_0)::OutputType ac_sqr_0_out[ceil_to_multiple<IndexType>(FC_0_OUTPUTS * 2, 32)];
- alignas(CacheLineSize) decltype(ac_0)::OutputBuffer ac_0_out;
- alignas(CacheLineSize) decltype(fc_1)::OutputBuffer fc_1_out;
- alignas(CacheLineSize) decltype(ac_1)::OutputBuffer ac_1_out;
- alignas(CacheLineSize) decltype(fc_2)::OutputBuffer fc_2_out;
-
- Buffer()
- {
- std::memset(this, 0, sizeof(*this));
- }
- };
+constexpr IndexType TransformedFeatureDimensions = 2560;
+constexpr IndexType PSQTBuckets = 8;
+constexpr IndexType LayerStacks = 8;
+
+struct Network {
+ static constexpr int FC_0_OUTPUTS = 15;
+ static constexpr int FC_1_OUTPUTS = 32;
+
+ Layers::AffineTransformSparseInput<TransformedFeatureDimensions, FC_0_OUTPUTS + 1> fc_0;
+ Layers::SqrClippedReLU<FC_0_OUTPUTS + 1> ac_sqr_0;
+ Layers::ClippedReLU<FC_0_OUTPUTS + 1> ac_0;
+ Layers::AffineTransform<FC_0_OUTPUTS * 2, FC_1_OUTPUTS> fc_1;
+ Layers::ClippedReLU<FC_1_OUTPUTS> ac_1;
+ Layers::AffineTransform<FC_1_OUTPUTS, 1> fc_2;
+
+ // Hash value embedded in the evaluation file
+ static constexpr std::uint32_t get_hash_value() {
+ // input slice hash
+ std::uint32_t hashValue = 0xEC42E90Du;
+ hashValue ^= TransformedFeatureDimensions * 2;
+
+ hashValue = decltype(fc_0)::get_hash_value(hashValue);
+ hashValue = decltype(ac_0)::get_hash_value(hashValue);
+ hashValue = decltype(fc_1)::get_hash_value(hashValue);
+ hashValue = decltype(ac_1)::get_hash_value(hashValue);
+ hashValue = decltype(fc_2)::get_hash_value(hashValue);
+
+ return hashValue;
+ }
+
+ // Read network parameters
+ bool read_parameters(std::istream& stream) {
+ return fc_0.read_parameters(stream) && ac_0.read_parameters(stream)
+ && fc_1.read_parameters(stream) && ac_1.read_parameters(stream)
+ && fc_2.read_parameters(stream);
+ }
+
+ // Write network parameters
+ bool write_parameters(std::ostream& stream) const {
+ return fc_0.write_parameters(stream) && ac_0.write_parameters(stream)
+ && fc_1.write_parameters(stream) && ac_1.write_parameters(stream)
+ && fc_2.write_parameters(stream);
+ }
+
+ std::int32_t propagate(const TransformedFeatureType* transformedFeatures) {
+ struct alignas(CacheLineSize) Buffer {
+ alignas(CacheLineSize) decltype(fc_0)::OutputBuffer fc_0_out;
+ alignas(CacheLineSize) decltype(ac_sqr_0)::OutputType
+ ac_sqr_0_out[ceil_to_multiple<IndexType>(FC_0_OUTPUTS * 2, 32)];
+ alignas(CacheLineSize) decltype(ac_0)::OutputBuffer ac_0_out;
+ alignas(CacheLineSize) decltype(fc_1)::OutputBuffer fc_1_out;
+ alignas(CacheLineSize) decltype(ac_1)::OutputBuffer ac_1_out;
+ alignas(CacheLineSize) decltype(fc_2)::OutputBuffer fc_2_out;
+
+ Buffer() { std::memset(this, 0, sizeof(*this)); }
+ };
#if defined(__clang__) && (__APPLE__)
- // workaround for a bug reported with xcode 12
- static thread_local auto tlsBuffer = std::make_unique<Buffer>();
- // Access TLS only once, cache result.
- Buffer& buffer = *tlsBuffer;
+ // workaround for a bug reported with xcode 12
+ static thread_local auto tlsBuffer = std::make_unique<Buffer>();
+ // Access TLS only once, cache result.
+ Buffer& buffer = *tlsBuffer;
#else
- alignas(CacheLineSize) static thread_local Buffer buffer;
+ alignas(CacheLineSize) static thread_local Buffer buffer;
#endif
- fc_0.propagate(transformedFeatures, buffer.fc_0_out);
- ac_sqr_0.propagate(buffer.fc_0_out, buffer.ac_sqr_0_out);
- ac_0.propagate(buffer.fc_0_out, buffer.ac_0_out);
- std::memcpy(buffer.ac_sqr_0_out + FC_0_OUTPUTS, buffer.ac_0_out, FC_0_OUTPUTS * sizeof(decltype(ac_0)::OutputType));
- fc_1.propagate(buffer.ac_sqr_0_out, buffer.fc_1_out);
- ac_1.propagate(buffer.fc_1_out, buffer.ac_1_out);
- fc_2.propagate(buffer.ac_1_out, buffer.fc_2_out);
-
- // buffer.fc_0_out[FC_0_OUTPUTS] is such that 1.0 is equal to 127*(1<<WeightScaleBits) in quantized form
- // but we want 1.0 to be equal to 600*OutputScale
- std::int32_t fwdOut = int(buffer.fc_0_out[FC_0_OUTPUTS]) * (600*OutputScale) / (127*(1<<WeightScaleBits));
- std::int32_t outputValue = buffer.fc_2_out[0] + fwdOut;
-
- return outputValue;
- }
+ fc_0.propagate(transformedFeatures, buffer.fc_0_out);
+ ac_sqr_0.propagate(buffer.fc_0_out, buffer.ac_sqr_0_out);
+ ac_0.propagate(buffer.fc_0_out, buffer.ac_0_out);
+ std::memcpy(buffer.ac_sqr_0_out + FC_0_OUTPUTS, buffer.ac_0_out,
+ FC_0_OUTPUTS * sizeof(decltype(ac_0)::OutputType));
+ fc_1.propagate(buffer.ac_sqr_0_out, buffer.fc_1_out);
+ ac_1.propagate(buffer.fc_1_out, buffer.ac_1_out);
+ fc_2.propagate(buffer.ac_1_out, buffer.fc_2_out);
+
+ // buffer.fc_0_out[FC_0_OUTPUTS] is such that 1.0 is equal to 127*(1<<WeightScaleBits) in
+ // quantized form, but we want 1.0 to be equal to 600*OutputScale
+ std::int32_t fwdOut =
+ int(buffer.fc_0_out[FC_0_OUTPUTS]) * (600 * OutputScale) / (127 * (1 << WeightScaleBits));
+ std::int32_t outputValue = buffer.fc_2_out[0] + fwdOut;
+
+ return outputValue;
+ }
};
} // namespace Stockfish::Eval::NNUE
-#endif // #ifndef NNUE_ARCHITECTURE_H_INCLUDED
+#endif // #ifndef NNUE_ARCHITECTURE_H_INCLUDED
projects / stockfish / blobdiff