Merge branch 'tidy-up-iproductwrtderivbase' into 'master'
Tidy-up CUDA implementation of IProductWRTDerivBase and add CUDA kernels with additional parallelism See merge request !72
Showing
- Operators/IProductWRTDerivBase/IProductWRTDerivBaseCUDA.cu 4 additions, 1 deletionOperators/IProductWRTDerivBase/IProductWRTDerivBaseCUDA.cu
- Operators/IProductWRTDerivBase/IProductWRTDerivBaseCUDA.hpp 181 additions, 134 deletionsOperators/IProductWRTDerivBase/IProductWRTDerivBaseCUDA.hpp
- Operators/IProductWRTDerivBase/IProductWRTDerivBaseCUDAKernels.cuh 357 additions, 323 deletions.../IProductWRTDerivBase/IProductWRTDerivBaseCUDAKernels.cuh
- Operators/IProductWRTDerivBase/IProductWRTDerivBaseStdMat.hpp 32 additions, 33 deletions...ators/IProductWRTDerivBase/IProductWRTDerivBaseStdMat.hpp
- tests/test_ipwrtderivbasecuda.cpp 20 additions, 30 deletionstests/test_ipwrtderivbasecuda.cpp
Loading
Please register or sign in to comment