|
MayaFlux 1.0.0
Digital-First Multimedia Processing Framework
|
Checking a set of tensors against what a network needs, and packing it into one float32 buffer. More...
#include "MayaFlux/Kakshya/NDData/Tensor.hpp"#include "MayaFlux/Kakshya/NDData/TensorConvert.hpp"#include "MayaFlux/Yantra/Executors/GpuDispatchCore.hpp"Go to the source code of this file.
Classes | |
| struct | MayaFlux::Yantra::TensorKernels::PackedWeights |
| The required tensors as float32 in one run, in the order they were required. More... | |
| struct | MayaFlux::Yantra::TensorKernels::TensorRequirement |
| One tensor a network needs. More... | |
Namespaces | |
| namespace | MayaFlux |
| Main namespace for the Maya Flux audio engine. | |
| namespace | MayaFlux::Yantra |
| namespace | MayaFlux::Yantra::TensorKernels |
| namespace | MayaFlux::Yantra::TensorKernels::detail |
Functions | |
| std::string | MayaFlux::Yantra::TensorKernels::check_tensors (const Kakshya::TensorSet &weights, std::span< const TensorRequirement > required) |
| Whether a set holds every required tensor with the required shape. | |
| size_t | MayaFlux::Yantra::TensorKernels::detail::element_count (const std::vector< uint64_t > &shape) |
| std::expected< PackedWeights, std::string > | MayaFlux::Yantra::TensorKernels::pack_weights (const Kakshya::TensorSet &weights, std::span< const TensorRequirement > required) |
| Convert the required tensors to float32 and lay them out in the order of required. | |
| std::string | MayaFlux::Yantra::TensorKernels::detail::shape_text (const std::vector< uint64_t > &shape) |
| void | MayaFlux::Yantra::TensorKernels::upload_packed_weights (GpuDispatchCore &context, uint32_t set, size_t binding, const PackedWeights &packed) |
| Allocate a shared buffer on a context and fill it with packed weights. | |
Checking a set of tensors against what a network needs, and packing it into one float32 buffer.
A network built from the tensor kernels lists the tensors it needs by name and shape, in the order its layers use them. check_tensors says whether a set holds them, pack_weights converts them to float32 and lays them out one after another, and upload_packed_weights puts the result on a context, where PackedWeights::address gives each tensor's device address for a stage.
Definition in file Yantra/Executors/TensorKernels/Weights.hpp.