![]() |
Deepity
Ultra-fast, SIMD-accelerated Predictive Coding
|
Go to the source code of this file.
Classes | |
| class | Deep::CPUBackend |
| CPU implementation of IComputeBackend: SIMD (via Activations.h) for elementwise ops, BLAS for GEMM. Graph capture/replay are no-ops (there's no async device queue to record). More... | |
Namespaces | |
| namespace | Deep |
| This file exists to provide both Tensor.h and IComputeBackend.h with device types without introducing circular dependencies. | |