NVIDIA DeepStream SDK API Reference4.0 Release |
Holds the initialization parameters required for the NvDsInferContext interface.
Definition at line 221 of file nvdsinfer_context.h.
Data Fields | |
unsigned int | uniqueID |
Unique identifier for the instance. More... | |
NvDsInferNetworkMode | networkMode |
Internal data format to be used by the inference engine. More... | |
char | protoFilePath [_PATH_MAX] |
Path to the prototxt file. More... | |
char | modelFilePath [_PATH_MAX] |
Path to the caffemodel file. More... | |
char | uffFilePath [_PATH_MAX] |
Path to the UFF model file. More... | |
char | onnxFilePath [_PATH_MAX] |
Path to the ONNX model file. More... | |
char | tltEncodedModelFilePath [_PATH_MAX] |
Path to the TLT encoded model file. More... | |
char | int8CalibrationFilePath [_PATH_MAX] |
Path to the INT8 calibration file. More... | |
NvDsInferDimsCHW | uffDimsCHW |
Input dimensions for the UFF model. More... | |
NvDsInferUffInputOrder | uffInputOrder |
Original input order for the UFF model. More... | |
char | uffInputBlobName [_MAX_STR_LENGTH] |
Name of the input layer for the UFF model. More... | |
char | tltModelKey [_MAX_STR_LENGTH] |
String key for decoding the TLT encoded model. More... | |
char | modelEngineFilePath [_PATH_MAX] |
Path to the serialized model engine file. More... | |
unsigned int | maxBatchSize |
Max number of frames that will be inferred together in a batch. More... | |
char | labelsFilePath [_PATH_MAX] |
Path to the labels file containing strings for the class labels. More... | |
char | meanImageFilePath [_PATH_MAX] |
Path to the mean image file (PPM format). More... | |
float | networkScaleFactor |
Normalization factor to scale the input pixels with. More... | |
NvDsInferFormat | networkInputFormat |
Network input format. More... | |
float | offsets [_MAX_CHANNELS] |
Per channel offsets for mean subtraction. More... | |
unsigned int | numOffsets |
NvDsInferNetworkType | networkType |
Type of the network. More... | |
int | useDBScan |
Boolean indicating if DBScan should be used for object clustering. More... | |
unsigned int | numDetectedClasses |
Number of classes detected by a detector network. More... | |
NvDsInferDetectionParams * | perClassDetectionParams |
Per class detection parameters. More... | |
float | classifierThreshold |
Minimum confidence threshold for classifier to consider a label valid. More... | |
float | segmentationThreshold |
char ** | outputLayerNames |
Array of output Layer Names. More... | |
unsigned int | numOutputLayers |
char | customLibPath [_PATH_MAX] |
Path to the library containing custom methods required to support the network. More... | |
char | customBBoxParseFuncName [_MAX_STR_LENGTH] |
Name of the custom bounding box function in the custom library. More... | |
char | customClassifierParseFuncName [_MAX_STR_LENGTH] |
Name of the custom classifier attribute parsing function in the custom library. More... | |
int | copyInputToHostBuffers |
Boolean indicating if input layer contents should be copied to host memories for access in the application. More... | |
unsigned int | gpuID |
ID of the GPU to run the inference on. More... | |
int | useDLA |
Boolean indicating if DLA should be used. More... | |
int | dlaCore |
DLA Core to use. More... | |
unsigned int | outputBufferPoolSize |
Number of sets of output buffers (host and device) to be allocated. More... | |
char | customNetworkConfigFilePath [_PATH_MAX] |
Path to the config file for custom network creation. More... | |
float _NvDsInferContextInitParams::classifierThreshold |
Minimum confidence threshold for classifier to consider a label valid.
Definition at line 303 of file nvdsinfer_context.h.
int _NvDsInferContextInitParams::copyInputToHostBuffers |
Boolean indicating if input layer contents should be copied to host memories for access in the application.
Definition at line 322 of file nvdsinfer_context.h.
char _NvDsInferContextInitParams::customBBoxParseFuncName[_MAX_STR_LENGTH] |
Name of the custom bounding box function in the custom library.
Definition at line 315 of file nvdsinfer_context.h.
char _NvDsInferContextInitParams::customClassifierParseFuncName[_MAX_STR_LENGTH] |
Name of the custom classifier attribute parsing function in the custom library.
Definition at line 318 of file nvdsinfer_context.h.
char _NvDsInferContextInitParams::customLibPath[_PATH_MAX] |
Path to the library containing custom methods required to support the network.
Definition at line 313 of file nvdsinfer_context.h.
char _NvDsInferContextInitParams::customNetworkConfigFilePath[_PATH_MAX] |
Path to the config file for custom network creation.
This can be used to store custom properties required by the custom network creation function.
Definition at line 337 of file nvdsinfer_context.h.
int _NvDsInferContextInitParams::dlaCore |
DLA Core to use.
Definition at line 330 of file nvdsinfer_context.h.
unsigned int _NvDsInferContextInitParams::gpuID |
ID of the GPU to run the inference on.
Definition at line 325 of file nvdsinfer_context.h.
char _NvDsInferContextInitParams::int8CalibrationFilePath[_PATH_MAX] |
Path to the INT8 calibration file.
Required only when using INT8 mode.
Definition at line 242 of file nvdsinfer_context.h.
char _NvDsInferContextInitParams::labelsFilePath[_PATH_MAX] |
Path to the labels file containing strings for the class labels.
Specifying the labels file is not mandatory. Refer to the custom models section of DeepStreamSDK documentation for the file format.
Definition at line 268 of file nvdsinfer_context.h.
unsigned int _NvDsInferContextInitParams::maxBatchSize |
Max number of frames that will be inferred together in a batch.
The number of input frames in a batch should be less than or equal to this number.
Definition at line 262 of file nvdsinfer_context.h.
char _NvDsInferContextInitParams::meanImageFilePath[_PATH_MAX] |
Path to the mean image file (PPM format).
Resolution of the file should be equal to the network input resolution.
Definition at line 273 of file nvdsinfer_context.h.
char _NvDsInferContextInitParams::modelEngineFilePath[_PATH_MAX] |
Path to the serialized model engine file.
When using model engine file, other parameters required for creating the model engine are ignored.
Definition at line 257 of file nvdsinfer_context.h.
char _NvDsInferContextInitParams::modelFilePath[_PATH_MAX] |
Path to the caffemodel file.
Definition at line 233 of file nvdsinfer_context.h.
NvDsInferFormat _NvDsInferContextInitParams::networkInputFormat |
Network input format.
Definition at line 279 of file nvdsinfer_context.h.
NvDsInferNetworkMode _NvDsInferContextInitParams::networkMode |
Internal data format to be used by the inference engine.
Definition at line 228 of file nvdsinfer_context.h.
float _NvDsInferContextInitParams::networkScaleFactor |
Normalization factor to scale the input pixels with.
Definition at line 276 of file nvdsinfer_context.h.
NvDsInferNetworkType _NvDsInferContextInitParams::networkType |
Type of the network.
Definition at line 289 of file nvdsinfer_context.h.
unsigned int _NvDsInferContextInitParams::numDetectedClasses |
Number of classes detected by a detector network.
Definition at line 296 of file nvdsinfer_context.h.
unsigned int _NvDsInferContextInitParams::numOffsets |
Definition at line 286 of file nvdsinfer_context.h.
unsigned int _NvDsInferContextInitParams::numOutputLayers |
Definition at line 309 of file nvdsinfer_context.h.
float _NvDsInferContextInitParams::offsets[_MAX_CHANNELS] |
Per channel offsets for mean subtraction.
This is an alternative to the mean image file. The number of offsets in the array should be exactly equalto the number of input channels.
Definition at line 285 of file nvdsinfer_context.h.
char _NvDsInferContextInitParams::onnxFilePath[_PATH_MAX] |
Path to the ONNX model file.
Definition at line 237 of file nvdsinfer_context.h.
unsigned int _NvDsInferContextInitParams::outputBufferPoolSize |
Number of sets of output buffers (host and device) to be allocated.
Definition at line 333 of file nvdsinfer_context.h.
char** _NvDsInferContextInitParams::outputLayerNames |
Array of output Layer Names.
Definition at line 308 of file nvdsinfer_context.h.
NvDsInferDetectionParams* _NvDsInferContextInitParams::perClassDetectionParams |
Per class detection parameters.
The size of the array should be equal to numDetectedClasses.
Definition at line 300 of file nvdsinfer_context.h.
char _NvDsInferContextInitParams::protoFilePath[_PATH_MAX] |
Path to the prototxt file.
Definition at line 231 of file nvdsinfer_context.h.
float _NvDsInferContextInitParams::segmentationThreshold |
Definition at line 305 of file nvdsinfer_context.h.
char _NvDsInferContextInitParams::tltEncodedModelFilePath[_PATH_MAX] |
Path to the TLT encoded model file.
Definition at line 239 of file nvdsinfer_context.h.
char _NvDsInferContextInitParams::tltModelKey[_MAX_STR_LENGTH] |
String key for decoding the TLT encoded model.
Definition at line 252 of file nvdsinfer_context.h.
NvDsInferDimsCHW _NvDsInferContextInitParams::uffDimsCHW |
Input dimensions for the UFF model.
Definition at line 245 of file nvdsinfer_context.h.
char _NvDsInferContextInitParams::uffFilePath[_PATH_MAX] |
Path to the UFF model file.
Definition at line 235 of file nvdsinfer_context.h.
char _NvDsInferContextInitParams::uffInputBlobName[_MAX_STR_LENGTH] |
Name of the input layer for the UFF model.
Definition at line 249 of file nvdsinfer_context.h.
NvDsInferUffInputOrder _NvDsInferContextInitParams::uffInputOrder |
Original input order for the UFF model.
Definition at line 247 of file nvdsinfer_context.h.
unsigned int _NvDsInferContextInitParams::uniqueID |
Unique identifier for the instance.
This can be used to identify the instance generating log and error messages.
Definition at line 225 of file nvdsinfer_context.h.
int _NvDsInferContextInitParams::useDBScan |
Boolean indicating if DBScan should be used for object clustering.
OpenCV groupRectangles will be used if set to false.
Definition at line 293 of file nvdsinfer_context.h.
int _NvDsInferContextInitParams::useDLA |
Boolean indicating if DLA should be used.
Definition at line 328 of file nvdsinfer_context.h.