Deepity
Ultra-fast, SIMD-accelerated Predictive Coding
Loading...
Searching...
No Matches
CUDABackend.h File Reference
#include <deepity/backend/IComputeBackend.h>
#include <cublas_v2.h>
#include <deepity/backend/Tensor.h>

Go to the source code of this file.

Classes

class  Deep::CUDABackend
 CUDA implementation of IComputeBackend: device kernels for elementwise ops, cuBLAS for GEMM, CUTLASS for the fused forward path, CUDA graphs for capture/replay. See src/backend/cuda/ for the implementation, split by concern to mirror this interface's own section grouping. More...

Namespaces

namespace  Deep
 This file exists to provide both Tensor.h and IComputeBackend.h with device types without introducing circular dependencies.