thek0tyara/llama-cpp-turboquant
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions 4

Working implementation with tests

Browse source
This commit is contained in:
Alejandro Saucedo 2021-03-07 08:00:19 +00:00
parent cf7d46cd23
commit f02b9d6915
21 changed files with 297 additions and 216 deletions
Download patch file Download diff file Expand all files Collapse all files

8
examples/array_multiplication/src/Main.cpp
View file

@ -14,9 +14,9 @@ int main()
kp::Manager mgr;
auto tensorInA = mgr.tensor<float>({ 2.0, 4.0, 6.0 });
auto tensorInB = mgr.tensor<float>({ 0.0, 1.0, 2.0 });
auto tensorOut = mgr.tensor<float>({ 0.0, 0.0, 0.0 });
auto tensorInA = mgr.tensor({ 2.0, 4.0, 6.0 });
auto tensorInB = mgr.tensor({ 0.0, 1.0, 2.0 });
auto tensorOut = mgr.tensor({ 0.0, 0.0, 0.0 });
std::string shader(R"(
// The version to use
@ -49,7 +49,7 @@ int main()
// prints "Output { 0 4 12 }"
std::cout<< "Output: { ";
for (const float& elem : tensorOut->data()) {
for (const float& elem : tensorOut->vector()) {
std::cout << elem << " ";
}
std::cout << "}" << std::endl;

21
examples/logistic_regression/src/Main.cpp
View file

@ -17,19 +17,19 @@ int main()
kp::Manager mgr;
auto xI = mgr.tensor<float>({ 0, 1, 1, 1, 1 });
auto xJ = mgr.tensor<float>({ 0, 0, 0, 1, 1 });
auto xI = mgr.tensor({ 0, 1, 1, 1, 1 });
auto xJ = mgr.tensor({ 0, 0, 0, 1, 1 });
auto y = mgr.tensor<float>({ 0, 0, 0, 1, 1 });
auto y = mgr.tensor({ 0, 0, 0, 1, 1 });
auto wIn = mgr.tensor<float>({ 0.001, 0.001 });
auto wOutI = mgr.tensor<float>({ 0, 0, 0, 0, 0 });
auto wOutJ = mgr.tensor<float>({ 0, 0, 0, 0, 0 });
auto wIn = mgr.tensor({ 0.001, 0.001 });
auto wOutI = mgr.tensor({ 0, 0, 0, 0, 0 });
auto wOutJ = mgr.tensor({ 0, 0, 0, 0, 0 });
auto bIn = mgr.tensor<float>({ 0 });
auto bOut = mgr.tensor<float>({ 0, 0, 0, 0, 0 });
auto bIn = mgr.tensor({ 0 });
auto bOut = mgr.tensor({ 0, 0, 0, 0, 0 });
auto lOut = mgr.tensor<float>({ 0, 0, 0, 0, 0 });
auto lOut = mgr.tensor({ 0, 0, 0, 0, 0 });
std::vector<std::shared_ptr<kp::Tensor>> params = { xI, xJ, y,
wIn, wOutI, wOutJ,
@ -40,7 +40,8 @@ int main()
(uint32_t*)(kp::shader_data::shaders_glsl_logisticregression_comp_spv
+ kp::shader_data::shaders_glsl_logisticregression_comp_spv_len));
std::shared_ptr<kp::Algorithm> algo = mgr.algorithm(params, spirv);
std::shared_ptr<kp::Algorithm> algo = mgr.algorithm(
params, spirv, kp::Workgroup({ 5 }), kp::Constants({ 5.0 }));
mgr.sequence()->eval<kp::OpTensorSyncDevice>(params);