modelRunInterferenceRNN
Parameters
| Type | Name | Default | Description |
|---|---|---|---|
| Integer | nInputs | 1 | Number of scalar inputs |
| Integer | nOutputs | 1 | Number of scalar outputs |
| Modelica.Units.SI.Time | samplePeriod | 0.1 | sampling period/interval |
| Integer | nHistoricElements | 10 | Number of elements from sampling steps for each input fed to the neural net |
| Boolean | continuous | false | |
| Boolean | useClaRaDelay | true | Switch between available delay types: Clara and MSL |
| Types.RNNFlatteningMethod | flatteningMethod | Types.RNNFlatteningMethod.OldFIrstInputSeq | |
| Boolean | returnSequences | false | |
| Integer | batchSize | 1 | |
| SMArtInt.Internal.SMArtIntClass | smartint | ||
| Integer | yFlatDim | if returnSequences then batchSize*nOutputs*nHistoricElements else batchSize*nOutputs |
Connectors
| Type | Name | Default | Description |
|---|---|---|---|
| Modelica.Blocks.Interfaces.RealInput[batchSize,nInputs] | u | ||
| Modelica.Blocks.Interfaces.RealOutput[batchSize,nOutputs] | y | ||
| Modelica.Blocks.Interfaces.RealOutput[if returnSequences then batchSize else 1,if returnSequences then nOutputs else 1,if returnSequences then nHistoricElements else 1] | ySequences | y_sequences |
Components
| Type | Name | Default | Description |
|---|---|---|---|
| Real[yFlatDim] | y_flat | ||
| Real[if returnSequences then batchSize else 1,if returnSequences then nOutputs else 1,if returnSequences then nHistoricElements else 1] | y_sequences | ||
| Real | finalFlatTensor | ||
| SubModels.RNNFlattenInput[batchSize] | flattenedHistory | ||
| SubModels.RNNDeflattenOutput[batchSize] | unflattenOutput |