NVIDIA Docs Hub NVIDIA Holoscan NVIDIA Holoscan SDK v3.3.0 Function holoscan::utils::set_activation_per_model

Function holoscan::utils::set_activation_per_model

Defined in File holoinfer_utils.hpp

Function Documentation

gxf_result_t holoscan::utils::set_activation_per_model(std::shared_ptr<HoloInfer::InferenceSpecs> &inference_specs, const HoloInfer::Mappings &activation_map, const std::vector<HoloInfer::ActivationSpec> &activation_specs, const std::string &module)

Setting up activation for each model by activation_map and ActivationSpec

Parameters

inference_specs – Inference spec that are needed to set the it’s activation_map.
activation_map – The predefined activation_map comes from the InferenceOp’s activation_map parameter.
activation_specs – Activation specs received from the input port model_activation_specs of InferenceOp.
module – Module that called for setting up activation specifications

Previous Function holoscan::utils::report_error

Next Function holoscan::utils::transmit_data_per_model(gxf_context_t&, const HoloInfer::MultiMappings&, HoloInfer::DataMap&, OutputContext&, std::vector<std::string>&, HoloInfer::DimType&, bool, bool, const nvidia::gxf::Handle<nvidia::gxf::Allocator>&, const std::string&, CudaStreamHandler&)