tensor_ops_kernels.hpp File Reference

Go to the source code of this file.

Namespaces
	bnmf_algs
	Main namespace for bnmf-algs library.
	bnmf_algs::cuda
	cuda namespace contains functions that operate on nvidia GPUs using CUDA routines.
	bnmf_algs::cuda::kernel
	Namespace containing CUDA kernel and device functions.

Functions
template<typename Scalar >
__global__ void	bnmf_algs::cuda::kernel::sum_tensor3D (cudaPitchedPtr tensor, Scalar *out, size_t out_pitch, size_t axis, size_t n_rows, size_t n_cols, size_t n_layers)
	Sum the given 3D tensor along the given axis and write the results to the corresponding index of the given 2D pitched matrix memory. More...