nnfw_session Struct Reference

Public Member Functions
	~nnfw_session ()
NNFW_STATUS	load_model_from_file (const char *package_file_path)
NNFW_STATUS	train_set_traininfo (const nnfw_train_info *info)
NNFW_STATUS	train_prepare ()
NNFW_STATUS	train_input_tensorinfo (uint32_t index, nnfw_tensorinfo *ti)
NNFW_STATUS	train_expected_tensorinfo (uint32_t index, nnfw_tensorinfo *ti)
NNFW_STATUS	train_set_input (uint32_t index, void *input)
NNFW_STATUS	train_set_expected (uint32_t index, void *expected)
NNFW_STATUS	train_set_output (uint32_t index, NNFW_TYPE type, void *buffer, size_t length)
NNFW_STATUS	train_run (bool update_weights)
NNFW_STATUS	train_get_loss (uint32_t index, float *loss)
NNFW_STATUS	train_export_circle (const char *path)
NNFW_STATUS	train_export_checkpoint (const char *path)
NNFW_STATUS	train_import_checkpoint (const char *path)

Static Public Member Functions
static NNFW_STATUS	create (nnfw_session **session)
	Factory method. It creates and initialize nnfw_session.

Detailed Description

Definition at line 68 of file onert-micro.cpp.

Constructor & Destructor Documentation

~nnfw_session()

nnfw_session::~nnfw_session

(

)

Definition at line 165 of file onert-micro.cpp.

{ delete _train_interpreter; }

Member Function Documentation

create()

NNFW_STATUS nnfw_session::create ( nnfw_session **  session )

static

Factory method. It creates and initialize nnfw_session.

Note

Use factory instead of constructor to get status

Definition at line 149 of file onert-micro.cpp.

{
  if (session == nullptr)
    return NNFW_STATUS_UNEXPECTED_NULL;
 
  auto new_session = std::unique_ptr<nnfw_session>(new nnfw_session());
  *session = new_session.release();
 
  if (*session == nullptr)
  {
    return NNFW_STATUS_ERROR;
  }
 
  return NNFW_STATUS_NO_ERROR;
}

References NNFW_STATUS_ERROR, NNFW_STATUS_NO_ERROR, and NNFW_STATUS_UNEXPECTED_NULL.

Referenced by nnfw_create_session().

load_model_from_file()

NNFW_STATUS nnfw_session::load_model_from_file

(

const char *

package_file_path

)

Definition at line 272 of file onert-micro.cpp.

{
  _model_buf = readFile(file_path);
  _config.model_ptr = _model_buf.data();
  _config.model_size = _model_buf.size();
  // load training info
  loadTrainingInfo(_config.model_ptr);
  // TODO: this import should start on nnfw_prepare if inference_interpreter is introduced
  _train_interpreter->importTrainModel(_config.model_ptr, _config);
  return NNFW_STATUS_NO_ERROR;
}

References onert_micro::OMTrainingInterpreter::importTrainModel(), onert_micro::OMConfig::model_ptr, onert_micro::OMConfig::model_size, NNFW_STATUS_NO_ERROR, and readFile().

train_expected_tensorinfo()

NNFW_STATUS nnfw_session::train_expected_tensorinfo	(	uint32_t	index,
		nnfw_tensorinfo *	ti
	)

train_export_checkpoint()

NNFW_STATUS nnfw_session::train_export_checkpoint

(

const char *

path

)

Definition at line 324 of file onert-micro.cpp.

{
  _train_interpreter->saveCheckpoint(_config, path);
  return NNFW_STATUS_NO_ERROR;
}

References NNFW_STATUS_NO_ERROR, and onert_micro::OMTrainingInterpreter::saveCheckpoint().

train_export_circle()

NNFW_STATUS nnfw_session::train_export_circle

(

const char *

path

)

Definition at line 318 of file onert-micro.cpp.

{
  _train_interpreter->saveModel(_config, path);
  return NNFW_STATUS_NO_ERROR;
}

References NNFW_STATUS_NO_ERROR, and onert_micro::OMTrainingInterpreter::saveModel().

train_get_loss()

NNFW_STATUS nnfw_session::train_get_loss	(	uint32_t	index,
		float *	loss
	)

Definition at line 370 of file onert-micro.cpp.

{
  onert_micro::OMMetrics m;
  switch (_config.training_context.loss)
  {
    case onert_micro::CROSS_ENTROPY:
      m = onert_micro::CROSS_ENTROPY_METRICS;
      break;
    default:
      m = onert_micro::CROSS_ENTROPY_METRICS;
      break;
  }
 
  _train_interpreter->evaluateMetric(_config, m, reinterpret_cast<void *>(loss),
                                     _config.training_context.batch_size);
  return NNFW_STATUS_NO_ERROR;
}

References onert_micro::OMTrainingContext::batch_size, onert_micro::CROSS_ENTROPY, onert_micro::CROSS_ENTROPY_METRICS, onert_micro::OMTrainingInterpreter::evaluateMetric(), onert_micro::OMTrainingContext::loss, m, NNFW_STATUS_NO_ERROR, and onert_micro::OMConfig::training_context.

train_import_checkpoint()

NNFW_STATUS nnfw_session::train_import_checkpoint

(

const char *

path

)

Definition at line 330 of file onert-micro.cpp.

{
  _train_interpreter->loadCheckpoint(_config, path);
  return NNFW_STATUS_NO_ERROR;
}

References onert_micro::OMTrainingInterpreter::loadCheckpoint(), and NNFW_STATUS_NO_ERROR.

train_input_tensorinfo()

NNFW_STATUS nnfw_session::train_input_tensorinfo	(	uint32_t	index,
		nnfw_tensorinfo *	ti
	)

train_prepare()

NNFW_STATUS nnfw_session::train_prepare

(

)

Definition at line 284 of file onert-micro.cpp.

{
  // TODO: Implement remaining jobs if inference_interpreter is introduced
  // maybe interpreter initialization ?
  return NNFW_STATUS_NO_ERROR;
}

References NNFW_STATUS_NO_ERROR.

train_run()

NNFW_STATUS nnfw_session::train_run

(

bool

update_weights

)

Definition at line 291 of file onert-micro.cpp.

{
  if (update_weights)
  {
    // TOOD: micro support update_weights ???
    // Here we use this flag for distinguish inference and train in trainaing interpreter
    _train_interpreter->trainSingleStep(_config);
    _config.training_context.num_epoch =
      _config.training_context.num_step / _config.training_context.batch_size + 1;
  }
  else
  {
    // TODO: support multiple input/output
    assert(outputbuf != nullptr);
    _train_interpreter->allocateInputs();
    float *allocated_input_data = (float *)_train_interpreter->getInputDataAt(0);
    float *user_input_data = (float *)_train_interpreter->getInputData(0);
    memcpy(allocated_input_data, user_input_data,
           sizeof(float) * _train_interpreter->getInputSizeAt(0));
    _train_interpreter->run(_config);
    float *calculated_ptr = (float *)_train_interpreter->getOutputDataAt(0);
    memcpy(outputbuf, calculated_ptr, sizeof(float) * _train_interpreter->getOutputSizeAt(0));
    _train_interpreter->reset();
  }
  return NNFW_STATUS_NO_ERROR;
}

References onert_micro::OMTrainingInterpreter::allocateInputs(), onert_micro::OMTrainingContext::batch_size, onert_micro::OMTrainingInterpreter::getInputData(), onert_micro::OMTrainingInterpreter::getInputDataAt(), onert_micro::OMTrainingInterpreter::getInputSizeAt(), onert_micro::OMTrainingInterpreter::getOutputDataAt(), onert_micro::OMTrainingInterpreter::getOutputSizeAt(), NNFW_STATUS_NO_ERROR, onert_micro::OMTrainingContext::num_epoch, onert_micro::OMTrainingContext::num_step, onert_micro::OMTrainingInterpreter::reset(), onert_micro::OMTrainingInterpreter::run(), onert_micro::OMConfig::training_context, and onert_micro::OMTrainingInterpreter::trainSingleStep().

train_set_expected()

NNFW_STATUS nnfw_session::train_set_expected	(	uint32_t	index,
		void *	expected
	)

Definition at line 344 of file onert-micro.cpp.

{
  _train_interpreter->setTarget((uint8_t *)expected, index);
  return NNFW_STATUS_NO_ERROR;
}

References NNFW_STATUS_NO_ERROR, and onert_micro::OMTrainingInterpreter::setTarget().

train_set_input()

NNFW_STATUS nnfw_session::train_set_input	(	uint32_t	index,
		void *	input
	)

Definition at line 337 of file onert-micro.cpp.

{
  _train_interpreter->setInput((uint8_t *)input, index);
  return NNFW_STATUS_NO_ERROR;
}

References NNFW_STATUS_NO_ERROR, and onert_micro::OMTrainingInterpreter::setInput().

train_set_output()

NNFW_STATUS nnfw_session::train_set_output	(	uint32_t	index,
		NNFW_TYPE	type,
		void *	buffer,
		size_t	length
	)

Definition at line 350 of file onert-micro.cpp.

{
  outputbuf = (uint8_t *)buffer;
  return NNFW_STATUS_NO_ERROR;
}

References NNFW_STATUS_NO_ERROR.

train_set_traininfo()

NNFW_STATUS nnfw_session::train_set_traininfo

(

const nnfw_train_info *

info

)

Definition at line 357 of file onert-micro.cpp.

{
  _config.training_context.learning_rate = info->learning_rate;
  _config.training_context.batch_size = info->batch_size;
  _config.training_context.optimizer =
    (info->opt == NNFW_TRAIN_OPTIMIZER_ADAM) ? onert_micro::ADAM : onert_micro::SGD;
  _config.training_context.beta = info->adam_opt.beta;
  _config.training_context.beta_squares = info->adam_opt.beta2;
  _config.training_context.beta = info->adam_opt.epsilon;
  _config.training_context.num_of_train_layers = info->num_trainble_ops;
  return NNFW_STATUS_NO_ERROR;
}

References onert_micro::ADAM, onert_micro::OMTrainingContext::batch_size, onert_micro::OMTrainingContext::beta, onert_micro::OMTrainingContext::beta_squares, info, onert_micro::OMTrainingContext::learning_rate, NNFW_STATUS_NO_ERROR, NNFW_TRAIN_OPTIMIZER_ADAM, onert_micro::OMTrainingContext::num_of_train_layers, onert_micro::OMTrainingContext::optimizer, onert_micro::SGD, and onert_micro::OMConfig::training_context.

---

The documentation for this struct was generated from the following file:

• onert-micro/onert-micro/src/api/onert-micro.cpp