org.deeplearning4j.nn.layers

Class FrozenLayer

java.lang.Object

org.deeplearning4j.nn.layers.wrapper.BaseWrapperLayer

org.deeplearning4j.nn.layers.FrozenLayer

All Implemented Interfaces:

Serializable, Cloneable, Layer, Model

public class FrozenLayer
extends BaseWrapperLayer

For purposes of transfer learning A frozen layers wraps another dl4j layer within it. The params of the layer within it are "frozen" or in other words held constant During the forward pass the frozen layer behaves as the layer within it would during test regardless of the training/test mode the network is in. Backprop is skipped since parameters are not be updated.

Author:

susaneraly

See Also:

Serialized Form

Nested Class Summary

Nested classes/interfaces inherited from interface org.deeplearning4j.nn.api.Layer

Layer.TrainingMode, Layer.Type

Field Summary

Fields inherited from class org.deeplearning4j.nn.layers.wrapper.BaseWrapperLayer

underlying

Constructor Summary

Constructors

Constructor and Description
FrozenLayer(Layer insideLayer)

Method Summary

Modifier and Type	Method and Description
org.nd4j.linalg.api.ndarray.INDArray	activate(boolean training, LayerWorkspaceMgr workspaceMgr) Perform forward pass and return the activations array with the last set input
org.nd4j.linalg.api.ndarray.INDArray	activate(org.nd4j.linalg.api.ndarray.INDArray input, boolean training, LayerWorkspaceMgr workspaceMgr) Perform forward pass and return the activations array with the specified input
void	applyConstraints(int iteration, int epoch) Apply any constraints to the model
org.nd4j.linalg.primitives.Pair<Gradient,org.nd4j.linalg.api.ndarray.INDArray>	backpropGradient(org.nd4j.linalg.api.ndarray.INDArray epsilon, LayerWorkspaceMgr workspaceMgr) Calculate the gradient relative to the error in the next layer
double	calcL1(boolean backpropOnlyParams) Calculate the l1 regularization term 0.0 if regularization is not used.
double	calcL2(boolean backpropOnlyParams) Calculate the l2 regularization term 0.0 if regularization is not used.
Layer	clone() Clone the layer
void	computeGradientAndScore(LayerWorkspaceMgr workspaceMgr) Update the score
void	fit() All models have a fit method
void	fit(org.nd4j.linalg.api.ndarray.INDArray data, LayerWorkspaceMgr workspaceMgr) Fit the model to the given data
Layer	getInsideLayer()
Gradient	gradient() Get the gradient.
org.nd4j.linalg.primitives.Pair<Gradient,Double>	gradientAndScore() Get the gradient and score
void	init() Init the model
protected String	layerId()
void	logTestMode(boolean training)
void	logTestMode(Layer.TrainingMode training)
void	setBackpropGradientsViewArray(org.nd4j.linalg.api.ndarray.INDArray gradients) Set the gradients array as a view of the full (backprop) network parameters NOTE: this is intended to be used internally in MultiLayerNetwork and ComputationGraph, not by users.
void	setCacheMode(CacheMode mode) This method sets given CacheMode for current layer
void	update(Gradient gradient) Update layer weights and biases with gradient change
void	update(org.nd4j.linalg.api.ndarray.INDArray gradient, String paramType) Perform one update applying the gradient

Methods inherited from class org.deeplearning4j.nn.layers.wrapper.BaseWrapperLayer

accumulateScore, addListeners, batchSize, clear, clearNoiseWeightParams, conf, feedForwardMaskArray, getEpochCount, getGradientsViewArray, getIndex, getInputMiniBatchSize, getIterationCount, getListeners, getMaskArray, getOptimizer, getParam, initParams, input, isPretrainLayer, numParams, numParams, params, paramTable, paramTable, score, setConf, setEpochCount, setIndex, setInput, setInputMiniBatchSize, setIterationCount, setListeners, setListeners, setMaskArray, setParam, setParams, setParamsViewArray, setParamTable, transpose, type, validateInput

Methods inherited from class java.lang.Object

equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Detail

FrozenLayer

public FrozenLayer(Layer insideLayer)

Method Detail

setCacheMode

public void setCacheMode(CacheMode mode)

Description copied from interface: Layer

This method sets given CacheMode for current layer

Specified by:

setCacheMode in interface Layer

Overrides:

setCacheMode in class BaseWrapperLayer

layerId

protected String layerId()

calcL2

public double calcL2(boolean backpropOnlyParams)

Description copied from interface: Layer

Calculate the l2 regularization term
0.0 if regularization is not used. Or 0.5 * l2Coeff * l2Magnitude otherwise.
Note that this does not divide by mini-batch size

Specified by:

calcL2 in interface Layer

Overrides:

calcL2 in class BaseWrapperLayer

Parameters:

backpropOnlyParams - If true: calculate L2 based on backprop params only. If false: calculate based on all params (including pretrain params, if any)

Returns:

the l2 regularization term for this layer.

calcL1

public double calcL1(boolean backpropOnlyParams)

Description copied from interface: Layer

Calculate the l1 regularization term
0.0 if regularization is not used. Or l1Coeff * l1Magnitude otherwise.
Note that this does not divide by mini-batch size

Specified by:

calcL1 in interface Layer

Overrides:

calcL1 in class BaseWrapperLayer

Parameters:

backpropOnlyParams - If true: calculate L1 based on backprop params only. If false: calculate based on all params (including pretrain params, if any)

Returns:

the l1 regularization term for this layer.

backpropGradient

public org.nd4j.linalg.primitives.Pair<Gradient,org.nd4j.linalg.api.ndarray.INDArray> backpropGradient(org.nd4j.linalg.api.ndarray.INDArray epsilon,
                                                                                                       LayerWorkspaceMgr workspaceMgr)

Description copied from interface: Layer

Calculate the gradient relative to the error in the next layer

Specified by:

backpropGradient in interface Layer

Overrides:

backpropGradient in class BaseWrapperLayer

Parameters:

epsilon - w^(L+1)*delta^(L+1). Or, equiv: dC/da, i.e., (dC/dz)*(dz/da) = dC/da, where C is cost function a=sigma(z) is activation.

workspaceMgr - Workspace manager

Returns:

Pair where Gradient is gradient for this layer, INDArray is epsilon (activation gradient) needed by next layer, but before element-wise multiply by sigmaPrime(z). So for standard feed-forward layer, if this layer is L, then return.getSecond() == dL/dIn = (w^(L)*(delta^(L))^T)^T. Note that the returned array should be placed in the ArrayType.ACTIVATION_GRAD workspace via the workspace manager

activate

public org.nd4j.linalg.api.ndarray.INDArray activate(boolean training,
                                                     LayerWorkspaceMgr workspaceMgr)

Description copied from interface: Layer

Perform forward pass and return the activations array with the last set input

Specified by:

activate in interface Layer

Overrides:

activate in class BaseWrapperLayer

Parameters:

training - training or test mode

workspaceMgr - Workspace manager

Returns:

the activation (layer output) of the last specified input. Note that the returned array should be placed in the ArrayType.ACTIVATIONS workspace via the workspace manager

activate

public org.nd4j.linalg.api.ndarray.INDArray activate(org.nd4j.linalg.api.ndarray.INDArray input,
                                                     boolean training,
                                                     LayerWorkspaceMgr workspaceMgr)

Description copied from interface: Layer

Perform forward pass and return the activations array with the specified input

Specified by:

activate in interface Layer

Overrides:

activate in class BaseWrapperLayer

Parameters:

input - the input to use

training - train or test mode

workspaceMgr - Workspace manager.

Returns:

Activations array. Note that the returned array should be placed in the ArrayType.ACTIVATIONS workspace via the workspace manager

clone

public Layer clone()

Description copied from interface: Layer

Clone the layer

Specified by:

clone in interface Layer

Overrides:

clone in class BaseWrapperLayer

Returns:

fit

public void fit()

Description copied from interface: Model

All models have a fit method

Specified by:

fit in interface Model

Overrides:

fit in class BaseWrapperLayer

update

public void update(Gradient gradient)

Description copied from interface: Model

Update layer weights and biases with gradient change

Specified by:

update in interface Model

Overrides:

update in class BaseWrapperLayer

update

public void update(org.nd4j.linalg.api.ndarray.INDArray gradient,
                   String paramType)

Description copied from interface: Model

Perform one update applying the gradient

Specified by:

update in interface Model

Overrides:

update in class BaseWrapperLayer

Parameters:

gradient - the gradient to apply

computeGradientAndScore

public void computeGradientAndScore(LayerWorkspaceMgr workspaceMgr)

Description copied from interface: Model

Update the score

Specified by:

computeGradientAndScore in interface Model

Overrides:

computeGradientAndScore in class BaseWrapperLayer

setBackpropGradientsViewArray

public void setBackpropGradientsViewArray(org.nd4j.linalg.api.ndarray.INDArray gradients)

Description copied from interface: Model

Set the gradients array as a view of the full (backprop) network parameters NOTE: this is intended to be used internally in MultiLayerNetwork and ComputationGraph, not by users.

Specified by:

setBackpropGradientsViewArray in interface Model

Overrides:

setBackpropGradientsViewArray in class BaseWrapperLayer

Parameters:

gradients - a 1 x nParams row vector that is a view of the larger (MLN/CG) gradients array

fit

public void fit(org.nd4j.linalg.api.ndarray.INDArray data,
                LayerWorkspaceMgr workspaceMgr)

Description copied from interface: Model

Fit the model to the given data

Specified by:

fit in interface Model

Overrides:

fit in class BaseWrapperLayer

Parameters:

data - the data to fit the model to

gradient

public Gradient gradient()

Description copied from interface: Model

Get the gradient. Note that this method will not calculate the gradient, it will rather return the gradient that has been computed before. For calculating the gradient, see Model#computeGradientAndScore() .

Specified by:

gradient in interface Model

Overrides:

gradient in class BaseWrapperLayer

Returns:

the gradient for this model, as calculated before

gradientAndScore

public org.nd4j.linalg.primitives.Pair<Gradient,Double> gradientAndScore()

Description copied from interface: Model

Get the gradient and score

Specified by:

gradientAndScore in interface Model

Overrides:

gradientAndScore in class BaseWrapperLayer

Returns:

the gradient and score

applyConstraints

public void applyConstraints(int iteration,
                             int epoch)

Description copied from interface: Model

Apply any constraints to the model

Specified by:

applyConstraints in interface Model

Overrides:

applyConstraints in class BaseWrapperLayer

init

public void init()

Init the model

Specified by:

init in interface Model

Overrides:

init in class BaseWrapperLayer

logTestMode

public void logTestMode(boolean training)

logTestMode

public void logTestMode(Layer.TrainingMode training)

getInsideLayer

public Layer getInsideLayer()