modelBaseNeuralNet
Information
This base class defines the parameter interface for all classes using the TfLite/ONNX interface. This class does not contain any evaluation call of a TfLite/ONNX model and therefore it should not be used.
Parameters
| Type | Name | Default | Description |
|---|
| Integer | nInputElements | product(inputSizes) | |
| Integer | nOutputElements | product(outputSizes) | |
| Selected Model |
| String | pathToAIModel | "" | Choose path to AI model |
| Tensor sizing |
| Integer | inputDimensions | 1 | Number of input dimension |
| Integer | inputSizes | {1} | Vector with size of tensor in each dimension |
| Integer | outputDimensions | 1 | Number of output dimension |
| Integer | outputSizes | {1} | Vector with size of tensor in each dimension |
| RNN Timing Settings |
| Boolean | stateful | false | Activate state handling for RNN with state in-/outputs |
| Real | samplePeriod | 0 | Fixed sample period for RNNs |
| Advanced › Performance |
| Integer | numberOfThreads | 1 | Number of threads used for inference (0: Number of CPU cores) |
| Advanced › TFLite |
| Boolean | useFlexOps | false | Activate to allow Tensorflow Flex Ops for tflite models |
| Advanced › ONNX |
| Boolean | useGPU | false | Activate to use a compatible CUDA GPU for inference |
| Integer | gpuDeviceID | 0 | CUDA device ID |
| SMArtInt.Internal.Types.ExecutionMode | executionMode | SMArtInt.Internal.Types.ExecutionMode.Sequential | Execution mode for run inference |
