Skip to content

Lily418/Cuda_Convolution

Repository files navigation

To build simply run make

The binary takes 2 parameters

The first indicates which implementation to run which can be 0 - CPU Implementation 1 - Naive GPU 2 - Optimised GPU

The second indicates which file to read the input vector from.

e.g. to run the CPU implementation on the small sample bin/release/smb 0 sample_small.txt

e.g. to run the Naive GPU version on the larger sample bin/release/smb 1 sample.txt

About

An optimised 1d convolution operator running on the GPU.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published