Go to the source code of this file.
|
| bool | buildEngineFromOnnx (const std::string &onnxFilePath, const std::string &engineFilePath, int batchSize, int numSensors, bool fp16Mode, PrettyLogger *logger=nullptr) |
| | Builds a TensorRT engine from an ONNX model and saves it to the specified file path. More...
|
| |
◆ TRT_MAJOR_VERSION
| #define TRT_MAJOR_VERSION 10 |
◆ TRT_MINOR_VERSION
| #define TRT_MINOR_VERSION 3 |
◆ TRT_PATCH_VERSION
| #define TRT_PATCH_VERSION 0 |
◆ TRTUniquePtr
◆ buildEngineFromOnnx()
| bool buildEngineFromOnnx |
( |
const std::string & |
onnxFilePath, |
|
|
const std::string & |
engineFilePath, |
|
|
int |
batchSize, |
|
|
int |
numSensors, |
|
|
bool |
fp16Mode, |
|
|
PrettyLogger * |
logger = nullptr |
|
) |
| |
Builds a TensorRT engine from an ONNX model and saves it to the specified file path.
Compatible with TensorRT 10.3.
- Parameters
-
| onnxFilePath | Path to the input ONNX model |
| engineFilePath | Path where the output engine should be saved |
| batchSize | Sparse4D model batch size |
| numSensors | Number of sensors [Deepstream batch size] |
| fp16Mode | Whether to enable FP16 mode for faster inference |
| logger | Optional logger for structured logging |
- Returns
- True if the engine was built successfully, false otherwise
◆ logger