Merge CUDA kernels with serial code #19
Labels
improving
Improving an already existing feature
performance
Regarding the performance of the library
rework
This entails a major rework of the code structure
Now that the code has been rearranged as a header only "library" with PR #18, once the CUDA implementation with have been improved (issue #3) it will be merged in the current library, thus having a single library instead of two (right now the two versions are separate).
The text was updated successfully, but these errors were encountered: