NeuralNetConfiguration.Builder (deeplearning4j-core 0.4-rc3.7 API)

java.lang.Object
- org.deeplearning4j.nn.conf.NeuralNetConfiguration.Builder

All Implemented Interfaces:: Cloneable

Enclosing class:: NeuralNetConfiguration

public static class NeuralNetConfiguration.Builder
extends Object
implements Cloneable

Field Summary

Fields
Modifier and Type	Field and Description
`protected String`	`activationFunction`
`protected double`	`biasInit`
`protected Distribution`	`dist`
`protected double`	`dropOut`
`protected Updater`	`updater`
`protected WeightInit`	`weightInit`

Constructor Summary

Constructors
Constructor and Description

Builder()

Constructors
Constructor and Description
`Builder()`

Method Summary

All Methods Instance Methods Concrete Methods Deprecated Methods
Modifier and Type	Method and Description
`NeuralNetConfiguration.Builder`	`activation(String activationFunction)` Activation function / neuron non-linearity Typical values include: "relu" (rectified linear), "tanh", "sigmoid", "softmax", "hardtanh", "leakyrelu", "maxout", "softsign", "softplus"
`NeuralNetConfiguration.Builder`	`adamMeanDecay(double adamMeanDecay)` Mean decay rate for Adam updater.
`NeuralNetConfiguration.Builder`	`adamVarDecay(double adamVarDecay)` Variance decay rate for Adam updater.
`NeuralNetConfiguration.Builder`	`biasInit(double biasInit)`
`NeuralNetConfiguration`	`build()` Return a configuration based on this builder
`NeuralNetConfiguration.Builder`	`clone()`
`NeuralNetConfiguration.Builder`	`constrainGradientToUnitNorm(boolean constrainGradientToUnitNorm)` Deprecated.
`NeuralNetConfiguration.Builder`	`dist(Distribution dist)` Distribution to sample initial weights from.
`NeuralNetConfiguration.Builder`	`dropOut(double dropOut)`
`NeuralNetConfiguration.Builder`	`gradientNormalization(GradientNormalization gradientNormalization)` Gradient normalization strategy.
`NeuralNetConfiguration.Builder`	`gradientNormalizationThreshold(double threshold)` Threshold for gradient normalization, only used for GradientNormalization.ClipL2PerLayer, GradientNormalization.ClipL2PerParamType, and GradientNormalization.ClipElementWiseAbsoluteValue Not used otherwise. L2 threshold for first two types of clipping, or absolute value threshold for last type of clipping.
`NeuralNetConfiguration.Builder`	`iterations(int numIterations)` Number of optimization iterations.
`NeuralNetConfiguration.Builder`	`l1(double l1)` L1 regularization coefficient.
`NeuralNetConfiguration.Builder`	`l2(double l2)` L2 regularization coefficient.
`NeuralNetConfiguration.Builder`	`layer(Layer layer)` Layer class.
`NeuralNetConfiguration.Builder`	`learningRate(double learningRate)` Learning rate.
`NeuralNetConfiguration.Builder`	`learningRateAfter(Map<Integer,Double> learningRateAfter)` Learning rate schedule.
`NeuralNetConfiguration.Builder`	`learningRateScoreBasedDecayRate(double lrScoreBasedDecay)` Rate to decrease learningRate by when the score stops improving.
`NeuralNetConfiguration.ListBuilder`	`list(int size)` Number of layers not including input.
`NeuralNetConfiguration.Builder`	`maxNumLineSearchIterations(int maxNumLineSearchIterations)` Maximum number of line search iterations.
`NeuralNetConfiguration.Builder`	`miniBatch(boolean miniBatch)` Process input as minibatch vs full dataset.
`NeuralNetConfiguration.Builder`	`minimize(boolean minimize)` Objective function to minimize or maximize cost function Default set to minimize true.
`NeuralNetConfiguration.Builder`	`momentum(double momentum)` Momentum rate.
`NeuralNetConfiguration.Builder`	`momentumAfter(Map<Integer,Double> momentumAfter)` Momentum schedule.
`NeuralNetConfiguration.Builder`	`optimizationAlgo(OptimizationAlgorithm optimizationAlgo)`
`NeuralNetConfiguration.Builder`	`regularization(boolean useRegularization)` Whether to use regularization (l1, l2, dropout, etc
`NeuralNetConfiguration.Builder`	`rho(double rho)` Ada delta coefficient
`NeuralNetConfiguration.Builder`	`rmsDecay(double rmsDecay)` Decay rate for RMSProp.
`NeuralNetConfiguration.Builder`	`schedules(boolean schedules)` Whether to use schedules, learningRateAfter and momentumAfter
`NeuralNetConfiguration.Builder`	`seed(int seed)` Random number generator seed.
`NeuralNetConfiguration.Builder`	`seed(long seed)` Random number generator seed.
`NeuralNetConfiguration.Builder`	`stepFunction(StepFunction stepFunction)` Step function to apply for back track line search.
`NeuralNetConfiguration.Builder`	`timeSeriesLength(int timeSeriesLength)` Deprecated.
`NeuralNetConfiguration.Builder`	`updater(Updater updater)` Gradient updater.
`NeuralNetConfiguration.Builder`	`useDropConnect(boolean useDropConnect)` Use drop connect: multiply the coefficients by a binomial sampling wrt the dropout probability
`NeuralNetConfiguration.Builder`	`weightInit(WeightInit weightInit)` Weight initialization scheme.

Methods inherited from class java.lang.Object
equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

- Field Detail
  - activationFunction
```
protected String activationFunction
```
  - weightInit
```
protected WeightInit weightInit
```
  - biasInit
```
protected double biasInit
```
  - dist
```
protected Distribution dist
```
  - dropOut
```
protected double dropOut
```
  - updater
```
protected Updater updater
```
- Constructor Detail
  - Builder
```
public Builder()
```
- Method Detail
  - timeSeriesLength
```
@Deprecated
public NeuralNetConfiguration.Builder timeSeriesLength(int timeSeriesLength)
```
    Deprecated.
    
    Deprecated. + * Time series length + * @param timeSeriesLength + * @return +
  - miniBatch
```
public NeuralNetConfiguration.Builder miniBatch(boolean miniBatch)
```
    Process input as minibatch vs full dataset. Default set to true.
  - useDropConnect
```
public NeuralNetConfiguration.Builder useDropConnect(boolean useDropConnect)
```
    Use drop connect: multiply the coefficients by a binomial sampling wrt the dropout probability
    
    Parameters:
    
    useDropConnect - whether to use drop connect or not
    
    Returns:
    
    the
  - minimize
```
public NeuralNetConfiguration.Builder minimize(boolean minimize)
```
    Objective function to minimize or maximize cost function Default set to minimize true.
  - maxNumLineSearchIterations
```
public NeuralNetConfiguration.Builder maxNumLineSearchIterations(int maxNumLineSearchIterations)
```
    Maximum number of line search iterations. Only applies for line search optimizers: Line Search SGD, Conjugate Gradient, LBFGS is NOT applicable for standard SGD
    
    Parameters:
    
    maxNumLineSearchIterations - > 0
    
    Returns:
  - layer
```
public NeuralNetConfiguration.Builder layer(Layer layer)
```
    Layer class.
  - stepFunction
```
public NeuralNetConfiguration.Builder stepFunction(StepFunction stepFunction)
```
    Step function to apply for back track line search. Only applies for line search optimizers: Line Search SGD, Conjugate Gradient, LBFGS Options: DefaultStepFunction (default), NegativeDefaultStepFunction GradientStepFunction (for SGD), NegativeGradientStepFunction
  - list
```
public NeuralNetConfiguration.ListBuilder list(int size)
```
    Number of layers not including input.
  - iterations
```
public NeuralNetConfiguration.Builder iterations(int numIterations)
```
    Number of optimization iterations.
  - seed
```
public NeuralNetConfiguration.Builder seed(int seed)
```
    Random number generator seed. Used for reproducability between runs
  - seed
```
public NeuralNetConfiguration.Builder seed(long seed)
```
    Random number generator seed. Used for reproducability between runs
  - optimizationAlgo
```
public NeuralNetConfiguration.Builder optimizationAlgo(OptimizationAlgorithm optimizationAlgo)
```
  - constrainGradientToUnitNorm
```
@Deprecated
public NeuralNetConfiguration.Builder constrainGradientToUnitNorm(boolean constrainGradientToUnitNorm)
```
    Deprecated.
    
    Deprecated. Use .gradientNormalization(GradientNormalization) instead
    
    See Also:
    
    GradientNormalization
  - regularization
```
public NeuralNetConfiguration.Builder regularization(boolean useRegularization)
```
    Whether to use regularization (l1, l2, dropout, etc
  - schedules
```
public NeuralNetConfiguration.Builder schedules(boolean schedules)
```
    Whether to use schedules, learningRateAfter and momentumAfter
  - clone
```
public NeuralNetConfiguration.Builder clone()
```
    Overrides:
    
    clone in class Object
  - activation
```
public NeuralNetConfiguration.Builder activation(String activationFunction)
```
    Activation function / neuron non-linearity Typical values include:
    "relu" (rectified linear), "tanh", "sigmoid", "softmax", "hardtanh", "leakyrelu", "maxout", "softsign", "softplus"
  - weightInit
```
public NeuralNetConfiguration.Builder weightInit(WeightInit weightInit)
```
    Weight initialization scheme.
    
    See Also:
    
    WeightInit
  - biasInit
```
public NeuralNetConfiguration.Builder biasInit(double biasInit)
```
  - dist
```
public NeuralNetConfiguration.Builder dist(Distribution dist)
```
    Distribution to sample initial weights from. Used in conjunction with .weightInit(WeightInit.DISTRIBUTION).
  - learningRate
```
public NeuralNetConfiguration.Builder learningRate(double learningRate)
```
    Learning rate. Defaults to 1e-1
  - learningRateAfter
```
public NeuralNetConfiguration.Builder learningRateAfter(Map<Integer,Double> learningRateAfter)
```
    Learning rate schedule. Map of the iteration to the learning rate to apply at that iteration.
  - learningRateScoreBasedDecayRate
```
public NeuralNetConfiguration.Builder learningRateScoreBasedDecayRate(double lrScoreBasedDecay)
```
    Rate to decrease learningRate by when the score stops improving. Learning rate is multiplied by this rate so ideally keep between 0 and 1.
  - l1
```
public NeuralNetConfiguration.Builder l1(double l1)
```
    L1 regularization coefficient.
  - l2
```
public NeuralNetConfiguration.Builder l2(double l2)
```
    L2 regularization coefficient.
  - dropOut
```
public NeuralNetConfiguration.Builder dropOut(double dropOut)
```
  - momentum
```
public NeuralNetConfiguration.Builder momentum(double momentum)
```
    Momentum rate.
  - momentumAfter
```
public NeuralNetConfiguration.Builder momentumAfter(Map<Integer,Double> momentumAfter)
```
    Momentum schedule. Map of the iteration to the momentum rate to apply at that iteration.
  - updater
```
public NeuralNetConfiguration.Builder updater(Updater updater)
```
    Gradient updater. For example, Updater.SGD for standard stochastic gradient descent, Updater.NESTEROV for Nesterov momentum, Updater.RSMPROP for RMSProp, etc.
    
    See Also:
    
    Updater
  - rho
```
public NeuralNetConfiguration.Builder rho(double rho)
```
    Ada delta coefficient
    
    Parameters:
    
    rho -
    
    Returns:
  - rmsDecay
```
public NeuralNetConfiguration.Builder rmsDecay(double rmsDecay)
```
    Decay rate for RMSProp. Only applies if using .updater(Updater.RMSPROP)
  - adamMeanDecay
```
public NeuralNetConfiguration.Builder adamMeanDecay(double adamMeanDecay)
```
    Mean decay rate for Adam updater. Only applies if using .updater(Updater.ADAM)
  - adamVarDecay
```
public NeuralNetConfiguration.Builder adamVarDecay(double adamVarDecay)
```
    Variance decay rate for Adam updater. Only applies if using .updater(Updater.ADAM)
  - gradientNormalization
```
public NeuralNetConfiguration.Builder gradientNormalization(GradientNormalization gradientNormalization)
```
    Gradient normalization strategy. Used to specify gradient renormalization, gradient clipping etc.
    
    Parameters:
    
    gradientNormalization - Type of normalization to use. Defaults to None.
    
    See Also:
    
    GradientNormalization
  - gradientNormalizationThreshold
```
public NeuralNetConfiguration.Builder gradientNormalizationThreshold(double threshold)
```
    Threshold for gradient normalization, only used for GradientNormalization.ClipL2PerLayer, GradientNormalization.ClipL2PerParamType, and GradientNormalization.ClipElementWiseAbsoluteValue
    Not used otherwise.
    L2 threshold for first two types of clipping, or absolute value threshold for last type of clipping.
  - build
```
public NeuralNetConfiguration build()
```
    Return a configuration based on this builder
    
    Returns:

Class NeuralNetConfiguration.Builder

Field Summary

Constructor Summary

Method Summary

Methods inherited from class java.lang.Object

Field Detail

activationFunction

weightInit

biasInit

dist

dropOut

updater

Constructor Detail

Builder

Method Detail

timeSeriesLength

miniBatch

useDropConnect

minimize

maxNumLineSearchIterations

layer

stepFunction

list

iterations

seed

seed

optimizationAlgo

constrainGradientToUnitNorm

regularization

schedules

clone

activation

weightInit

biasInit

dist

learningRate

learningRateAfter

learningRateScoreBasedDecayRate

l1

l2

dropOut

momentum

momentumAfter

updater

rho

rmsDecay

adamMeanDecay

adamVarDecay

gradientNormalization

gradientNormalizationThreshold

build