Configuration struct for Torch edge classifier.
More...
#include <ActsPlugins/Gnn/TorchEdgeClassifier.hpp>
|
|
float | cut = 0.5 |
| | Classification score threshold for edge filtering.
|
|
int | deviceID = 0 |
| | CUDA device ID to use for inference.
|
|
std::string | modelPath |
| | Path to the PyTorch model file.
|
|
int | nChunks = 1 |
| | Number of chunks to process.
|
|
std::vector< int > | selectedFeatures = {} |
| | Selected feature indices for input.
|
|
bool | undirected = false |
| | Whether to treat graph as undirected.
|
|
bool | useEdgeFeatures = false |
| | Whether to use edge features.
|
Configuration struct for Torch edge classifier.