Package org.deeplearning4j.nn.conf.constraint

Class MinMaxNormConstraint

java.lang.Object

org.deeplearning4j.nn.conf.constraint.BaseConstraint

org.deeplearning4j.nn.conf.constraint.MinMaxNormConstraint

All Implemented Interfaces:

Serializable, Cloneable, LayerConstraint

public class MinMaxNormConstraint
extends BaseConstraint

See Also:

Serialized Form

Field Summary

Fields

Modifier and Type	Field	Description
static double	DEFAULT_RATE

Fields inherited from class org.deeplearning4j.nn.conf.constraint.BaseConstraint

DEFAULT_EPSILON, dimensions, epsilon, params

Constructor Summary

Constructors

Constructor	Description
MinMaxNormConstraint(double min, double max, double rate, int... dimensions)	Apply to weights but not biases by default
MinMaxNormConstraint(double min, double max, double rate, Set<String> paramNames, int... dimensions)
MinMaxNormConstraint(double min, double max, int... dimensions)	Apply to weights but not biases by default

Method Summary

Modifier and Type	Method	Description
void	apply(INDArray param)
MinMaxNormConstraint	clone()

Methods inherited from class org.deeplearning4j.nn.conf.constraint.BaseConstraint

applyConstraint, getBroadcastDims

Methods inherited from class java.lang.Object

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

Methods inherited from interface org.deeplearning4j.nn.api.layers.LayerConstraint

getParams, setParams

Field Detail

DEFAULT_RATE

public static final double DEFAULT_RATE

See Also:

Constant Field Values

Constructor Detail

MinMaxNormConstraint

public MinMaxNormConstraint(double min,
                            double max,
                            int... dimensions)

Apply to weights but not biases by default

Parameters:

max - Maximum L2 value

min - Minimum L2 value

dimensions - Dimensions to apply to. For DenseLayer, OutputLayer, RnnOutputLayer, LSTM, etc: this should be dimension 1. For CNNs, this should be dimensions [1,2,3] corresponding to last 3 of parameters which have order [depthOut, depthIn, kH, kW]

MinMaxNormConstraint

public MinMaxNormConstraint(double min,
                            double max,
                            double rate,
                            int... dimensions)

Apply to weights but not biases by default

Parameters:

max - Maximum L2 value

min - Minimum L2 value

rate - Constraint rate

dimensions - Dimensions to apply to. For DenseLayer, OutputLayer, RnnOutputLayer, LSTM, etc: this should be dimension 1. For CNNs, this should be dimensions [1,2,3] corresponding to last 3 of parameters which have order [depthOut, depthIn, kH, kW]

MinMaxNormConstraint

public MinMaxNormConstraint(double min,
                            double max,
                            double rate,
                            Set<String> paramNames,
                            int... dimensions)

Parameters:

max - Maximum L2 value

min - Minimum L2 value

rate - Constraint rate

paramNames - Which parameter names to apply constraint to

dimensions - Dimensions to apply to. For DenseLayer, OutputLayer, RnnOutputLayer, LSTM, etc: this should be dimension 1. For CNNs, this should be dimensions [1,2,3] corresponding to last 3 of parameters which have order [depthOut, depthIn, kH, kW]

Method Detail

apply

public void apply(INDArray param)

Specified by:

apply in class BaseConstraint

clone

public MinMaxNormConstraint clone()

Specified by:

clone in interface LayerConstraint

Specified by:

clone in class BaseConstraint