Class Acts::OnnxMetricLearning

class OnnxMetricLearning : public Acts::GraphConstructionBase

Public Functions

OnnxMetricLearning(const Config &cfg, std::unique_ptr<const Logger> logger)
inline Config config() const
virtual std::tuple<std::any, std::any> operator()(std::vector<float> &inputValues, std::size_t numNodes) override

Perform the graph construction.

  • inputValues – Flattened input data

  • numNodes – number of nodes. inputValues.size() / numNodes then gives the number of features


(node_tensor, edge_tensore)

struct Config

Public Members

int embeddingDim = 8
int knnVal = 500
std::string modelPath
float rVal = 1.6
int spacepointFeatures = 3