Class GaussianNoise
- java.lang.Object
-
- org.deeplearning4j.nn.conf.dropout.GaussianNoise
-
- All Implemented Interfaces:
Serializable
,Cloneable
,IDropout
public class GaussianNoise extends Object implements IDropout
- See Also:
- Serialized Form
-
-
Constructor Summary
Constructors Modifier Constructor Description GaussianNoise(double stddev)
protected
GaussianNoise(double stddev, ISchedule stddevSchedule)
GaussianNoise(ISchedule stddevSchedule)
-
Method Summary
All Methods Instance Methods Concrete Methods Modifier and Type Method Description INDArray
applyDropout(INDArray inputActivations, INDArray output, int iteration, int epoch, LayerWorkspaceMgr workspaceMgr)
INDArray
backprop(INDArray gradAtOutput, INDArray gradAtInput, int iteration, int epoch)
Perform backprop.void
clear()
Clear the internal state (for example, dropout mask) if any is presentIDropout
clone()
-
-
-
Constructor Detail
-
GaussianNoise
public GaussianNoise(double stddev)
- Parameters:
stddev
- Standard deviation for the mean 0 Gaussian noise
-
GaussianNoise
public GaussianNoise(ISchedule stddevSchedule)
- Parameters:
stddevSchedule
- Schedule for standard deviation for the mean 0 Gaussian noise
-
GaussianNoise
protected GaussianNoise(double stddev, ISchedule stddevSchedule)
-
-
Method Detail
-
applyDropout
public INDArray applyDropout(INDArray inputActivations, INDArray output, int iteration, int epoch, LayerWorkspaceMgr workspaceMgr)
- Specified by:
applyDropout
in interfaceIDropout
- Parameters:
inputActivations
- Input activations arrayoutput
- The result array (same as inputArray for in-place ops) for the post-dropout activationsiteration
- Current iteration numberepoch
- Current epoch numberworkspaceMgr
- Workspace manager, if any storage is required (use ArrayType.INPUT)- Returns:
- The output (resultArray) after applying dropout
-
backprop
public INDArray backprop(INDArray gradAtOutput, INDArray gradAtInput, int iteration, int epoch)
Description copied from interface:IDropout
Perform backprop. This should also clear the internal state (dropout mask) if any is present- Specified by:
backprop
in interfaceIDropout
- Parameters:
gradAtOutput
- Gradients at the output of the dropout op - i.e., dL/dOutgradAtInput
- Gradients at the input of the dropout op - i.e., dL/dIn. Use the same array as gradAtOutput to apply the backprop gradient in-placeiteration
- Current iterationepoch
- Current epoch- Returns:
- Same array as gradAtInput - i.e., gradient after backpropagating through dropout op - i.e., dL/dIn
-
clear
public void clear()
Description copied from interface:IDropout
Clear the internal state (for example, dropout mask) if any is present
-
-