MayaFlux 1.0.0
Digital-First Multimedia Processing Framework
Loading...
Searching...
No Matches

◆ SileroVadModel() [2/4]

MayaFlux::Yantra::SileroVadModel::SileroVadModel ( const Kakshya::TensorSet & weights,
std::shared_ptr< GpuDispatchRecorder > recording )

As above, appending to a recording the caller shares with other work.

With a shared recording the network and a consumer such as the interval extractor go to the device in one submission. The recording must be compute isolated.

Parameters
weightsA set that diagnose accepts.
recordingRecording the model records into. A null recording gives the model its own.

Definition at line 221 of file SileroVadModel.cpp.

222{
223 auto packed = TensorKernels::pack_weights(weights, slot_specs());
224 if (!packed) {
226 std::source_location::current(), "SileroVadModel: {}", packed.error());
227 }
228 m_impl = std::make_unique<Impl>(std::move(*packed), std::move(recording));
229}
@ ComputeMatrix
Compute operations (Yantra - algorithms, matrices, DSP).
@ Yantra
DSP algorithms, computational units, matrix operations, Grammar.
std::expected< PackedWeights, std::string > 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.

References MayaFlux::Journal::ComputeMatrix, m_impl, MayaFlux::Yantra::TensorKernels::pack_weights(), and MayaFlux::Journal::Yantra.

Here is the call graph for this function: