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

Updated python to build

Browse source 
Signed-off-by: Alejandro Saucedo <axsauze@gmail.com>
This commit is contained in:
Alejandro Saucedo 2021-09-12 13:13:20 +01:00
parent 5193975e5c
commit 6113d286a9
2 changed files with 48 additions and 1 deletions
Download patch file Download diff file Expand all files Collapse all files

1
python/src/main.cpp
View file

@ -61,7 +61,6 @@ PYBIND11_MODULE(kp, m) {
py::class_<kp::Algorithm, std::shared_ptr<kp::Algorithm>>(m, "Algorithm", DOC(kp, Algorithm, Algorithm))
.def("get_tensors", &kp::Algorithm::getTensors, DOC(kp, Algorithm, getTensors))
.def("destroy", &kp::Algorithm::destroy, DOC(kp, Algorithm, destroy))
.def("get_spec_consts", &kp::Algorithm::getSpecializationConstants, DOC(kp, Algorithm, getSpecializationConstants))
.def("is_init", &kp::Algorithm::isInit, DOC(kp, Algorithm, isInit));
py::class_<kp::Tensor, std::shared_ptr<kp::Tensor>>(m, "Tensor", DOC(kp, Tensor))

48
test/TestSpecializationConstant.cpp
View file

@ -53,3 +53,51 @@ TEST(TestSpecializationConstants, TestTwoConstants)
}
}
}
TEST(TestSpecializationConstants, TestConstantsInt)
{
{
std::string shader(R"(
#version 450
layout (constant_id = 0) const float cOne = 1;
layout (constant_id = 1) const float cTwo = 1;
layout (local_size_x = 1) in;
layout(set = 0, binding = 0) buffer a { float pa[]; };
layout(set = 0, binding = 1) buffer b { float pb[]; };
void main() {
uint index = gl_GlobalInvocationID.x;
pa[index] = cOne;
pb[index] = cTwo;
})");
std::vector<uint32_t> spirv = compileSource(shader);
std::shared_ptr<kp::Sequence> sq = nullptr;
{
kp::Manager mgr;
std::shared_ptr<kp::TensorT<float>> tensorA =
mgr.tensor({ 0, 0, 0 });
std::shared_ptr<kp::TensorT<float>> tensorB =
mgr.tensor({ 0, 0, 0 });
std::vector<std::shared_ptr<kp::Tensor>> params = { tensorA,
tensorB };
kp::Constants spec = kp::Constants({ 5.0, 0.3 });
std::shared_ptr<kp::Algorithm> algo =
mgr.algorithm(params, spirv, {}, spec);
sq = mgr.sequence()
->record<kp::OpTensorSyncDevice>(params)
->record<kp::OpAlgoDispatch>(algo)
->record<kp::OpTensorSyncLocal>(params)
->eval();
EXPECT_EQ(tensorA->vector(), std::vector<float>({ 5, 5, 5 }));
EXPECT_EQ(tensorB->vector(), std::vector<float>({ 0.3, 0.3, 0.3 }));
}
}
}