Package deepboof.impl.backward.standard
Class DFunctionDropOut_F64
java.lang.Object
deepboof.impl.forward.standard.BaseFunction<T>
deepboof.impl.backward.standard.BaseDFunction<deepboof.tensors.Tensor_F64>
deepboof.impl.backward.standard.DFunctionDropOut_F64
- All Implemented Interfaces:
DFunctionDropOut<deepboof.tensors.Tensor_F64>,deepboof.DFunction<deepboof.tensors.Tensor_F64>,deepboof.Function<deepboof.tensors.Tensor_F64>
public class DFunctionDropOut_F64 extends BaseDFunction<deepboof.tensors.Tensor_F64> implements DFunctionDropOut<deepboof.tensors.Tensor_F64>
Implementation of
DFunctionDropOut for Tensor_F64-
Field Summary
Fields inherited from class deepboof.impl.backward.standard.BaseDFunction
learningModeFields inherited from class deepboof.impl.forward.standard.BaseFunction
miniBatchSize, parameters, shapeInput, shapeOutput, shapeParameters -
Constructor Summary
Constructors Constructor Description DFunctionDropOut_F64(long randomSeed, double dropRate)Configures drop out -
Method Summary
Modifier and Type Method Description protected void_backwards(deepboof.tensors.Tensor_F64 input, deepboof.tensors.Tensor_F64 dout, deepboof.tensors.Tensor_F64 gradientInput, List<deepboof.tensors.Tensor_F64> gradientParameters)void_forward(deepboof.tensors.Tensor_F64 input, deepboof.tensors.Tensor_F64 output)void_initialize()void_setParameters(List<deepboof.tensors.Tensor_F64> parameters)doublegetDropRate()Returns a number from 0 to 1 indicating the likelihood of a neuron being dropped.Class<deepboof.tensors.Tensor_F64>getTensorType()Methods inherited from class deepboof.impl.backward.standard.BaseDFunction
backwards, evaluating, isLearning, learningMethods inherited from class deepboof.impl.forward.standard.BaseFunction
forward, getOutputShape, getParameters, getParameterShapes, initialize, setParametersMethods inherited from class java.lang.Object
clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, waitMethods inherited from interface deepboof.DFunction
backwards, evaluating, isLearning, learningMethods inherited from interface deepboof.Function
forward, getOutputShape, getParameters, getParameterShapes, initialize, setParameters
-
Constructor Details
-
DFunctionDropOut_F64
public DFunctionDropOut_F64(long randomSeed, double dropRate)Configures drop out- Parameters:
randomSeed- random seed used to pick which neurons are droppeddropRate- Fraction of time a neuron is dropped
-
-
Method Details
-
_initialize
public void _initialize()- Specified by:
_initializein classdeepboof.impl.forward.standard.BaseFunction<deepboof.tensors.Tensor_F64>
-
_setParameters
- Specified by:
_setParametersin classdeepboof.impl.forward.standard.BaseFunction<deepboof.tensors.Tensor_F64>
-
_forward
public void _forward(deepboof.tensors.Tensor_F64 input, deepboof.tensors.Tensor_F64 output)- Specified by:
_forwardin classdeepboof.impl.forward.standard.BaseFunction<deepboof.tensors.Tensor_F64>
-
getDropRate
public double getDropRate()Description copied from interface:DFunctionDropOutReturns a number from 0 to 1 indicating the likelihood of a neuron being dropped. 0 = 0% change and 1 = 100% chance- Specified by:
getDropRatein interfaceDFunctionDropOut<deepboof.tensors.Tensor_F64>- Returns:
- drop rate
-
_backwards
protected void _backwards(deepboof.tensors.Tensor_F64 input, deepboof.tensors.Tensor_F64 dout, deepboof.tensors.Tensor_F64 gradientInput, List<deepboof.tensors.Tensor_F64> gradientParameters)- Specified by:
_backwardsin classBaseDFunction<deepboof.tensors.Tensor_F64>
-
getTensorType
- Specified by:
getTensorTypein interfacedeepboof.Function<deepboof.tensors.Tensor_F64>
-