File runner.hpp¶
-
namespace xir¶
-
namespace vart¶
-
template<typename InputType, typename OutputType = InputType>
class BaseRunner Subclassed by Runner
Public Functions
-
virtual ~BaseRunner() = default¶
-
virtual std::pair<std::uint32_t, int> execute_async(InputType input, OutputType output) = 0
-
- Parameters:
input – inputs with a customized type
output – outputs with a customized type
- Returns:
pair<jobid, status> status 0 for exit successfully, others for customized warnings or errors
-
virtual int wait(int jobid, int timeout = -1) = 0
wait
modes: 1. Blocking wait for specific ID. 2. Non-blocking wait for specific ID. 3. Blocking wait for any ID. 4. Non-blocking wait for any ID
- Parameters:
jobid – job id, neg for any id, others for specific job id
timeout – timeout, neg for block for ever, 0 for non-block, pos for block with a limitation(ms).
- Returns:
status 0 for exit successfully, others for customized warnings or errors
-
virtual ~BaseRunner() = default¶
-
class Runner : public vart::BaseRunner<const std::vector<TensorBuffer*>&>
- #include <runner.hpp>
Class of the Runner, provides API to use the runner.
The runner instance has a number of member functions to control the execution and get the input and output tensors of the runner.
Sample code:
// This example assumes that you have a DPU subgraph called dpu_subgraph. // The way to create a DPU runner to run dpu_subgraph is shown below. // create runner auto runner = vart::Runner::create_runner(dpu_subgraph, ”run”); // get input tensors auto input_tensors = runner->get_input_tensors(); // get input tensor buffers auto input_tensor_buffers = std::vector<vart::TensorBuffer*>(); for (auto input : input_tensors) { auto t = vart::alloc_cpu_flat_tensor_buffer(input); input_tensor_buffers.emplace_back(t.get()); } // get output tensors auto output_tensors = runner->get_output_tensors(); // get output tensor buffers auto output_tensor_buffers = std::vector< vart::TensorBuffer*>(); for (auto output : output _tensors) { auto t = vart::alloc_cpu_flat_tensor_buffer(output); output_tensor_buffers.emplace_back(t.get()); } // sync input tensor buffers for (auto& input : input_tensor_buffers) { input->sync_for_write(0, input->get_tensor()->get_data_size() / input->get_tensor()->get_shape()[0]); } // run runner auto v = runner->execute_async(input_tensor_buffers, output_tensor_buffers); auto status = runner->wait((int)v.first, 1000000000); // sync output tensor buffers for (auto& output : output_tensor_buffers) { output->sync_for_read(0, output->get_tensor()->get_data_size() / output->get_tensor()->get_shape()[0]); }
Subclassed by RunnerExt
Public Functions
-
virtual ~Runner() = default¶
-
virtual std::pair<uint32_t, int> execute_async(const std::vector<TensorBuffer*> &input, const std::vector<TensorBuffer*> &output) = 0
Executes the runner.
This is a blocking function.
- Parameters:
input – A vector of TensorBuffer create by all input tensors of runner.
output – A vector of TensorBuffer create by all output tensors of runner.
- Returns:
pair<jobid, status> status 0 for exit successfully, others for customized warnings or errors
-
virtual int wait(int jobid, int timeout) = 0
Waits for the end of DPU processing.
modes: 1. Blocking wait for specific ID. 2. Non-blocking wait for specific ID. 3. Blocking wait for any ID. 4. Non-blocking wait for any ID
- Parameters:
jobid – job id, neg for any id, others for specific job id
timeout – timeout, neg for block for ever, 0 for non-block, pos for block with a limitation(ms).
- Returns:
status 0 for exit successfully, others for customized warnings or errors
-
virtual TensorFormat get_tensor_format()
Get the tensor format of runner.
Sample code:
auto format = runner->get_tensor_format(); switch (format) { case vart::Runner::TensorFormat::NCHW: // do something break; case vart::Runner::TensorFormat::NHWC: // do something break; }
- Returns:
TensorFormat : NHWC / HCHW
-
virtual std::vector<const xir::Tensor*> get_input_tensors() = 0
Get all input tensors of runner.
Sample code:
inputTensors = runner->get_input_tensors(); for (auto input : inputTensor) { input->get_name(); input->get_shape(); input->get_element_num(); }
- Returns:
All input tensors. A vector of raw pointer to the input tensor.
-
virtual std::vector<const xir::Tensor*> get_output_tensors() = 0
Get all output tensors of runner.
Sample code:
outputTensors = runner->get_output_tensors(); for (auto output : outputTensor) { output->get_name(); output->get_shape(); output->get_element_num(); }
- Returns:
All output tensors. A vector of raw pointer to the output tensor.
-
virtual std::pair<std::uint32_t, int> execute_async(InputType input, OutputType output) = 0
-
- Parameters:
input – inputs with a customized type
output – outputs with a customized type
- Returns:
pair<jobid, status> status 0 for exit successfully, others for customized warnings or errors
Public Static Functions
-
static std::unique_ptr<Runner> create_runner(const xir::Subgraph *subgraph, const std::string &mode = std::string(""))
Factory function to create an instance of DPU runner by subgraph.
Sample code:
// This API can be used like: auto runner = vart::Runner::create_runner(subgraph, "run");
- Parameters:
subgraph – XIR Subgraph
mode – 1 mode supported: ‘run’ - DPU runner.
- Returns:
An instance of DPU runner.
-
static std::unique_ptr<Runner> create_runner_with_attrs(const xir::Subgraph *subgraph, xir::Attrs *attrs)
Factory function to create an instance of DPU runner by subgraph, and attrs.
- Parameters:
subgraph – XIR Subgraph
attrs – XIR attrs object, this object is shared among all runners on the same graph.
attrs["mode"], 1 – mode supported: ‘run’ - DPU runner.
- Returns:
An instance of DPU runner.
-
static std::vector<std::unique_ptr<Runner>> create_runner(const std::string &model_directory)
-
virtual ~Runner() = default¶
-
template<typename InputType, typename OutputType = InputType>