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

Updated tensor to copy to memory explicitly, assessing why copy doesn't last through operations

Browse source
This commit is contained in:
Alejandro Saucedo 2020-08-22 18:08:56 +01:00
parent 6f0203b863
commit e047aa3b43
10 changed files with 65 additions and 79 deletions
Download patch file Download diff file Expand all files Collapse all files

4
src/Algorithm.cpp
View file

@ -41,6 +41,10 @@ Algorithm::init(std::string shaderFilePath,
this->createPipeline();
}
void Algorithm::createDescriptorPool() {
}
void
Algorithm::createParameters(std::vector<std::shared_ptr<Tensor>>& tensorParams)
{