*/
#include "RecordMinMax.h"
-#include "RecordFunction.h"
#include "MinMaxObserver.h"
#include <luci/Importer.h>
}
}
-void update_quantparam(record_minmax::MinMaxObserver *observer, const std::string &mode,
- float min_percentile, float max_percentile)
-{
- auto minmax_map = observer->minMaxData()->getMap();
- for (auto iter = minmax_map->begin(); iter != minmax_map->end(); ++iter)
- {
- auto node = iter->first;
- auto minmax = iter->second;
-
- float min{0.0f}, max{0.0f};
- if (mode == "percentile")
- {
- min = record_minmax::getNthPercentile(minmax.min_vector, min_percentile);
- max = record_minmax::getNthPercentile(minmax.max_vector, max_percentile);
- }
- else if (mode == "moving_average")
- {
- min = record_minmax::getMovingAverage(minmax.min_vector, 0.9, 16, true);
- max = record_minmax::getMovingAverage(minmax.max_vector, 0.9, 16, false);
- }
- assert(mode == "percentile" || mode == "moving_average");
- auto quantparam = std::make_unique<luci::CircleQuantParam>();
- quantparam->min.push_back(min);
- quantparam->max.push_back(max);
-
- assert(node->quantparam() == nullptr);
-
- auto mutable_node = const_cast<luci::CircleNode *>(node);
- mutable_node->quantparam(std::move(quantparam));
- }
-}
-
} // namespace
namespace record_minmax
// The directory should contain binary files each of which is a raw data,
// ready to be consumed by the input circle model without any modification
// TODO reduce duplicate codes with profileRawData
-void RecordMinMax::profileRawDataDirectory(const std::string &mode,
- const std::string &input_data_path, float min_percentile,
- float max_percentile)
+void RecordMinMax::profileRawDataDirectory(const std::string &input_data_path)
{
struct dirent *entry = nullptr;
DIR *dp = nullptr;
std::cout << "Recording finished. Number of recorded data: " << num_records << std::endl;
- update_quantparam(getObserver(), mode, min_percentile, max_percentile);
+ _minmax_computer->update_qparam(getObserver()->minMaxData()->getMap());
}
// input_data_path is a text file which specifies the representative data
// ready to be consumed by the input circle model without any modification
// NOTE If a model has multiple inputs, the binary file should have inputs concatenated in the same
// order with the input index of the circle model.
-void RecordMinMax::profileRawData(const std::string &mode, const std::string &input_data_path,
- float min_percentile, float max_percentile)
+void RecordMinMax::profileRawData(const std::string &input_data_path)
{
std::ifstream input_file(input_data_path);
if (input_file.fail())
std::cout << "Recording finished. Number of recorded data: " << num_records << std::endl;
- update_quantparam(getObserver(), mode, min_percentile, max_percentile);
+ _minmax_computer->update_qparam(getObserver()->minMaxData()->getMap());
}
WholeOutput RecordMinMax::importH5Data(const std::string &input_data_path)
{
DataType dtype;
Shape shape;
- importer.readTensor(i, input_idx, &dtype, &shape, input_data.data());
+ importer.readTensor(i, input_idx, &dtype, &shape, input_data.data(), input_data.size());
// Check the type and the shape of the input data is valid
verifyTypeShape(input_node, dtype, shape);
else
{
// Skip type/shape check for raw data
- importer.readTensor(i, input_idx, input_data.data());
+ importer.readTensor(i, input_idx, input_data.data(), input_data.size());
}
whole_output[i].emplace_back(std::move(input_data));
}
}
}
-void RecordMinMax::profileData(const std::string &mode, const std::string &input_data_path,
- float min_percentile, float max_percentile)
+void RecordMinMax::profileData(const std::string &input_data_path)
{
try
{
{
DataType dtype;
Shape shape;
- importer.readTensor(record_idx, input_idx, &dtype, &shape, input_data.data());
+ importer.readTensor(record_idx, input_idx, &dtype, &shape, input_data.data(),
+ input_data.size());
// Check the type and the shape of the input data is valid
verifyTypeShape(input_node, dtype, shape);
else
{
// Skip type/shape check for raw data
- importer.readTensor(record_idx, input_idx, input_data.data());
+ importer.readTensor(record_idx, input_idx, input_data.data(), input_data.size());
}
// TODO: Input data is copied twice (file -> buffer (input_data) -> interpreter inputs)
throw std::runtime_error("HDF5 error occurred.");
}
- update_quantparam(getObserver(), mode, min_percentile, max_percentile);
+ _minmax_computer->update_qparam(getObserver()->minMaxData()->getMap());
}
-void RecordMinMax::profileDataInParallel(const std::string &mode,
- const std::string &input_data_path, float min_percentile,
- float max_percentile)
+void RecordMinMax::profileDataInParallel(const std::string &input_data_path)
{
LOGGER(l);
// End parallel part
- // Copy all min, max values to one observer
- auto observer = std::make_unique<MinMaxObserver>();
- auto main_min_max_map = const_cast<MinMaxMap *>(observer->minMaxData());
+ // Copy all min, max values to one min/max map
+ MinMaxMap main_min_max_map;
for (const auto &obs : _observers)
{
const auto node = iter.first;
const auto &minmax = iter.second;
- main_min_max_map->appendMinMaxVector(node, minmax);
+ main_min_max_map.appendMinMaxVector(node, minmax);
}
}
std::cout << "Recording finished. Number of recorded data: " << num_records << std::endl;
- update_quantparam(observer.get(), mode, min_percentile, max_percentile);
+ _minmax_computer->update_qparam(main_min_max_map.getMap());
}
-void RecordMinMax::profileDataWithRandomInputs(const std::string &mode, float min_percentile,
- float max_percentile)
+void RecordMinMax::profileDataWithRandomInputs(void)
{
// We use three randomly-generated records
const uint32_t num_records = 3;
std::cout << "Recording finished. Number of recorded data: " << num_records << std::endl;
- update_quantparam(getObserver(), mode, min_percentile, max_percentile);
+ _minmax_computer->update_qparam(getObserver()->minMaxData()->getMap());
}
void RecordMinMax::saveModel(const std::string &output_model_path)
projects / platform / core / ml / nnfw.git / blobdiff