FineTuneConfiguration.Builder (deeplearning4j-nn 1.0.0-beta2 API)

java.lang.Object
- org.deeplearning4j.nn.transferlearning.FineTuneConfiguration.Builder

Enclosing class:: FineTuneConfiguration

public static class FineTuneConfiguration.Builder
extends Object

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
`FineTuneConfiguration.Builder`	`activation(Activation activation)` Activation function / neuron non-linearity
`FineTuneConfiguration.Builder`	`activation(IActivation activationFn)` Activation function / neuron non-linearity
`FineTuneConfiguration.Builder`	`backprop(boolean backprop)`
`FineTuneConfiguration.Builder`	`backpropType(BackpropType backpropType)` The type of backprop.
`FineTuneConfiguration.Builder`	`biasInit(double biasInit)` Constant for bias initialization.
`FineTuneConfiguration.Builder`	`biasUpdater(IUpdater biasUpdater)` Gradient updater configuration, for the biases only.
`FineTuneConfiguration`	`build()`
`FineTuneConfiguration.Builder`	`constraints(List<LayerConstraint> constraints)` Set constraints to be applied to all layers.
`FineTuneConfiguration.Builder`	`convolutionMode(ConvolutionMode convolutionMode)` Sets the convolution mode for convolutional layers, which impacts padding and output sizes.
`FineTuneConfiguration.Builder`	`cudnnAlgoMode(ConvolutionLayer.AlgoMode cudnnAlgoMode)` Sets the cuDNN algo mode for convolutional layers, which impacts performance and memory usage of cuDNN.
`FineTuneConfiguration.Builder`	`dist(Distribution dist)` Distribution to sample initial weights from.
`FineTuneConfiguration.Builder`	`dropOut(double inputRetainProbability)` Dropout probability.
`FineTuneConfiguration.Builder`	`dropout(IDropout dropout)` Set the dropout
`FineTuneConfiguration.Builder`	`gradientNormalization(GradientNormalization gradientNormalization)` Gradient normalization strategy.
`FineTuneConfiguration.Builder`	`gradientNormalizationThreshold(double gradientNormalizationThreshold)` 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
`FineTuneConfiguration.Builder`	`inferenceWorkspaceMode(WorkspaceMode inferenceWorkspaceMode)` This method defines Workspace mode being used during inference: NONE: workspace won't be used ENABLED: workspaces will be used for inference (reduced memory and better performance)
`FineTuneConfiguration.Builder`	`l1(double l1)` L1 regularization coefficient for the weights
`FineTuneConfiguration.Builder`	`l1Bias(double l1Bias)` L1 regularization coefficient for the bias parameters
`FineTuneConfiguration.Builder`	`l2(double l2)` L2 regularization coefficient for the weights
`FineTuneConfiguration.Builder`	`l2Bias(double l2Bias)` L2 regularization coefficient for the bias parameters
`FineTuneConfiguration.Builder`	`maxNumLineSearchIterations(int maxNumLineSearchIterations)`
`FineTuneConfiguration.Builder`	`miniBatch(boolean miniBatch)` Whether scores and gradients should be divided by the minibatch size. Most users should leave this ast he default value of true.
`FineTuneConfiguration.Builder`	`minimize(boolean minimize)`
`FineTuneConfiguration.Builder`	`optimizationAlgo(OptimizationAlgorithm optimizationAlgo)`
`FineTuneConfiguration.Builder`	`pretrain(boolean pretrain)`
`FineTuneConfiguration.Builder`	`seed(int seed)` RNG seed for reproducibility
`FineTuneConfiguration.Builder`	`seed(long seed)` RNG seed for reproducibility
`FineTuneConfiguration.Builder`	`stepFunction(StepFunction stepFunction)`
`FineTuneConfiguration.Builder`	`tbpttBackLength(int tbpttBackLength)` When doing truncated BPTT: how many steps of backward should we do? Only applicable when doing backpropType(BackpropType.TruncatedBPTT) This is the k2 parameter on pg23 of http://www.cs.utoronto.ca/~ilya/pubs/ilya_sutskever_phd_thesis.pdf
`FineTuneConfiguration.Builder`	`tbpttFwdLength(int tbpttFwdLength)` When doing truncated BPTT: how many steps of forward pass should we do before doing (truncated) backprop? Only applicable when doing backpropType(BackpropType.TruncatedBPTT) Typically tBPTTForwardLength parameter is same as the tBPTTBackwardLength parameter, but may be larger than it in some circumstances (but never smaller) Ideally your training data time series length should be divisible by this This is the k1 parameter on pg23 of http://www.cs.utoronto.ca/~ilya/pubs/ilya_sutskever_phd_thesis.pdf
`String`	`toString()`
`FineTuneConfiguration.Builder`	`trainingWorkspaceMode(WorkspaceMode trainingWorkspaceMode)` This method defines Workspace mode being used during training: NONE: workspace won't be used ENABLED: workspaces will be used for training (reduced memory and better performance)
`FineTuneConfiguration.Builder`	`updater(IUpdater updater)` Gradient updater configuration.
`FineTuneConfiguration.Builder`	`updater(Updater updater)` Deprecated. Use `updater(IUpdater)`
`FineTuneConfiguration.Builder`	`weightInit(Distribution distribution)` Set weight initialization scheme to random sampling via the specified distribution. Equivalent to: `.weightInit(WeightInit.DISTRIBUTION).dist(distribution)`
`FineTuneConfiguration.Builder`	`weightInit(WeightInit weightInit)` Weight initialization scheme
`FineTuneConfiguration.Builder`	`weightNoise(IWeightNoise weightNoise)` Set the weight noise (such as `DropConnect` and `WeightNoise`)

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

- Constructor Detail
  - Builder
```
public Builder()
```
- Method Detail
  - activation
```
public FineTuneConfiguration.Builder activation(IActivation activationFn)
```
    Activation function / neuron non-linearity
  - activation
```
public FineTuneConfiguration.Builder activation(Activation activation)
```
    Activation function / neuron non-linearity
  - weightInit
```
public FineTuneConfiguration.Builder weightInit(WeightInit weightInit)
```
    Weight initialization scheme
    
    See Also:
    
    WeightInit
  - weightInit
```
public FineTuneConfiguration.Builder weightInit(Distribution distribution)
```
    Set weight initialization scheme to random sampling via the specified distribution.
    Equivalent to: .weightInit(WeightInit.DISTRIBUTION).dist(distribution)
    
    Parameters:
    
    distribution - Distribution to use for weight initialization
  - biasInit
```
public FineTuneConfiguration.Builder biasInit(double biasInit)
```
    Constant for bias initialization. Default: 0.0
    
    Parameters:
    
    biasInit - Constant for bias initialization
  - dist
```
public FineTuneConfiguration.Builder dist(Distribution dist)
```
    Distribution to sample initial weights from. Used in conjunction with .weightInit(WeightInit.DISTRIBUTION)
    
    See Also:
    
    weightInit(Distribution)
  - l1
```
public FineTuneConfiguration.Builder l1(double l1)
```
    L1 regularization coefficient for the weights
  - l2
```
public FineTuneConfiguration.Builder l2(double l2)
```
    L2 regularization coefficient for the weights
  - l1Bias
```
public FineTuneConfiguration.Builder l1Bias(double l1Bias)
```
    L1 regularization coefficient for the bias parameters
  - l2Bias
```
public FineTuneConfiguration.Builder l2Bias(double l2Bias)
```
    L2 regularization coefficient for the bias parameters
  - dropout
```
public FineTuneConfiguration.Builder dropout(IDropout dropout)
```
    Set the dropout
    
    Parameters:
    
    dropout - Dropout, such as Dropout, GaussianDropout, GaussianNoise etc
  - dropOut
```
public FineTuneConfiguration.Builder dropOut(double inputRetainProbability)
```
    Dropout probability. This is the probability of retaining each input activation value for a layer. dropOut(x) will keep an input activation with probability x, and set to 0 with probability 1-x.
    dropOut(0.0) is a special value / special case - when set to 0.0., dropout is disabled (not applied). Note that a dropout value of 1.0 is functionally equivalent to no dropout: i.e., 100% probability of retaining each input activation.
    
    Note 1: Dropout is applied at training time only - and is automatically not applied at test time (for evaluation, etc)
    Note 2: This sets the probability per-layer. Care should be taken when setting lower values for complex networks (too much information may be lost with aggressive (very low) dropout values).
    Note 3: Frequently, dropout is not applied to (or, has higher retain probability for) input (first layer) layers. Dropout is also often not applied to output layers. This needs to be handled MANUALLY by the user - set .dropout(0) on those layers when using global dropout setting.
    Note 4: Implementation detail (most users can ignore): DL4J uses inverted dropout, as described here: http://cs231n.github.io/neural-networks-2/
    
    Parameters:
    
    inputRetainProbability - Dropout probability (probability of retaining each input activation value for a layer)
    
    See Also:
    
    dropout(IDropout)
  - weightNoise
```
public FineTuneConfiguration.Builder weightNoise(IWeightNoise weightNoise)
```
    Set the weight noise (such as DropConnect and WeightNoise)
    
    Parameters:
    
    weightNoise - Weight noise instance to use
  - updater
```
public FineTuneConfiguration.Builder updater(IUpdater updater)
```
    Gradient updater configuration. For example, Adam or Nesterovs
    
    Parameters:
    
    updater - Updater to use
  - updater
```
@Deprecated
public FineTuneConfiguration.Builder updater(Updater updater)
```
    Deprecated. Use updater(IUpdater)
  - biasUpdater
```
public FineTuneConfiguration.Builder biasUpdater(IUpdater biasUpdater)
```
    Gradient updater configuration, for the biases only. If not set, biases will use the updater as set by updater(IUpdater)
    
    Parameters:
    
    biasUpdater - Updater to use for bias parameters
  - miniBatch
```
public FineTuneConfiguration.Builder miniBatch(boolean miniBatch)
```
    Whether scores and gradients should be divided by the minibatch size.
    Most users should leave this ast he default value of true.
  - maxNumLineSearchIterations
```
public FineTuneConfiguration.Builder maxNumLineSearchIterations(int maxNumLineSearchIterations)
```
  - seed
```
public FineTuneConfiguration.Builder seed(long seed)
```
    RNG seed for reproducibility
    
    Parameters:
    
    seed - RNG seed to use
  - seed
```
public FineTuneConfiguration.Builder seed(int seed)
```
    RNG seed for reproducibility
    
    Parameters:
    
    seed - RNG seed to use
  - optimizationAlgo
```
public FineTuneConfiguration.Builder optimizationAlgo(OptimizationAlgorithm optimizationAlgo)
```
  - stepFunction
```
public FineTuneConfiguration.Builder stepFunction(StepFunction stepFunction)
```
  - minimize
```
public FineTuneConfiguration.Builder minimize(boolean minimize)
```
  - gradientNormalization
```
public FineTuneConfiguration.Builder gradientNormalization(GradientNormalization gradientNormalization)
```
    Gradient normalization strategy. Used to specify gradient renormalization, gradient clipping etc. See GradientNormalization for details
    
    Parameters:
    
    gradientNormalization - Type of normalization to use. Defaults to None.
    
    See Also:
    
    GradientNormalization
  - gradientNormalizationThreshold
```
public FineTuneConfiguration.Builder gradientNormalizationThreshold(double gradientNormalizationThreshold)
```
    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
  - convolutionMode
```
public FineTuneConfiguration.Builder convolutionMode(ConvolutionMode convolutionMode)
```
    Sets the convolution mode for convolutional layers, which impacts padding and output sizes. See ConvolutionMode for details. Defaults to ConvolutionMode.TRUNCATE
    
    Parameters:
    
    convolutionMode - Convolution mode to use
  - cudnnAlgoMode
```
public FineTuneConfiguration.Builder cudnnAlgoMode(ConvolutionLayer.AlgoMode cudnnAlgoMode)
```
    Sets the cuDNN algo mode for convolutional layers, which impacts performance and memory usage of cuDNN. See ConvolutionLayer.AlgoMode for details. Defaults to "PREFER_FASTEST", but "NO_WORKSPACE" uses less memory.
  - constraints
```
public FineTuneConfiguration.Builder constraints(List<LayerConstraint> constraints)
```
    Set constraints to be applied to all layers. Default: no constraints.
    Constraints can be used to enforce certain conditions (non-negativity of parameters, max-norm regularization, etc). These constraints are applied at each iteration, after the parameters have been updated.
    
    Parameters:
    
    constraints - Constraints to apply to all parameters of all layers
  - pretrain
```
public FineTuneConfiguration.Builder pretrain(boolean pretrain)
```
  - backprop
```
public FineTuneConfiguration.Builder backprop(boolean backprop)
```
  - backpropType
```
public FineTuneConfiguration.Builder backpropType(BackpropType backpropType)
```
    The type of backprop. Default setting is used for most networks (MLP, CNN etc), but optionally truncated BPTT can be used for training recurrent neural networks. If using TruncatedBPTT make sure you set both tBPTTForwardLength() and tBPTTBackwardLength()
    
    Parameters:
    
    backpropType - Type of backprop. Default: BackpropType.Standard
  - tbpttFwdLength
```
public FineTuneConfiguration.Builder tbpttFwdLength(int tbpttFwdLength)
```
    When doing truncated BPTT: how many steps of forward pass should we do before doing (truncated) backprop?
    Only applicable when doing backpropType(BackpropType.TruncatedBPTT)
    Typically tBPTTForwardLength parameter is same as the tBPTTBackwardLength parameter, but may be larger than it in some circumstances (but never smaller)
    Ideally your training data time series length should be divisible by this This is the k1 parameter on pg23 of http://www.cs.utoronto.ca/~ilya/pubs/ilya_sutskever_phd_thesis.pdf
    
    Parameters:
    
    tbpttFwdLength - Forward length > 0, >= backwardLength
  - tbpttBackLength
```
public FineTuneConfiguration.Builder tbpttBackLength(int tbpttBackLength)
```
    When doing truncated BPTT: how many steps of backward should we do?
    Only applicable when doing backpropType(BackpropType.TruncatedBPTT)
    This is the k2 parameter on pg23 of http://www.cs.utoronto.ca/~ilya/pubs/ilya_sutskever_phd_thesis.pdf
    
    Parameters:
    
    tbpttBackLength - <= forwardLength
  - trainingWorkspaceMode
```
public FineTuneConfiguration.Builder trainingWorkspaceMode(WorkspaceMode trainingWorkspaceMode)
```
    This method defines Workspace mode being used during training: NONE: workspace won't be used ENABLED: workspaces will be used for training (reduced memory and better performance)
    
    Parameters:
    
    trainingWorkspaceMode - Workspace mode for training
    
    Returns:
    
    Builder
  - inferenceWorkspaceMode
```
public FineTuneConfiguration.Builder inferenceWorkspaceMode(WorkspaceMode inferenceWorkspaceMode)
```
    This method defines Workspace mode being used during inference:
    NONE: workspace won't be used
    ENABLED: workspaces will be used for inference (reduced memory and better performance)
    
    Parameters:
    
    inferenceWorkspaceMode - Workspace mode for inference
    
    Returns:
    
    Builder
  - build
```
public FineTuneConfiguration build()
```
  - toString
```
public String toString()
```
    Overrides:
    
    toString in class Object

Class FineTuneConfiguration.Builder

Constructor Summary

Method Summary

Methods inherited from class java.lang.Object

Constructor Detail

Builder

Method Detail

activation

activation

weightInit

weightInit

biasInit

dist

l1

l2

l1Bias

l2Bias

dropout

dropOut

weightNoise

updater

updater

biasUpdater

miniBatch

maxNumLineSearchIterations

seed

seed

optimizationAlgo

stepFunction

minimize

gradientNormalization

gradientNormalizationThreshold

convolutionMode

cudnnAlgoMode

constraints

pretrain

backprop

backpropType

tbpttFwdLength

tbpttBackLength

trainingWorkspaceMode

inferenceWorkspaceMode

build

toString