gdTrainParameters {ganGenerativeData} | R Documentation |
Specify parameters for training of generative model
Description
Specify parameters for training of neural networks used for generation of generative data. These parameters are passed to function gdTrain().
Usage
gdTrainParameters(
numberOfTrainingIterations = 10000,
numberOfInitializationIterations = 1500,
numberOfHiddenLayerUnits = 1024,
learningRate = 7e-05,
dropout = 0.05
)
Arguments
numberOfTrainingIterations |
Number of training iterations |
numberOfInitializationIterations |
Number of initialization iterations |
numberOfHiddenLayerUnits |
Number of hidden layer units |
learningRate |
Learning rate for training of neural networks |
dropout |
Value in the range of 0 to 1. Specifies the rate of hidden units that are dropped. Dropout is a regularization method to prevent overfitting. |
Value
List of parameters for training of generative model
Examples
## Not run:
generateParameters <- gdGenerateParameters(numberOfTrainingIterations = 10000)
## End(Not run)
[Package ganGenerativeData version 2.0.2 Index]