MultilayerPerceptron

java.lang.Object
- smile.base.mlp.MultilayerPerceptron

All Implemented Interfaces:

java.io.Serializable

Direct Known Subclasses:

MLP, MLP
```
public abstract class MultilayerPerceptron
extends java.lang.Object
implements java.io.Serializable
```
Fully connected multilayer perceptron neural network. An MLP consists of at least three layers of nodes: an input layer, a hidden layer and an output layer. The nodes are interconnected through weighted acyclic arcs from each preceding layer to the following, without lateral or feedback connections. Each node calculates a transformed weighted linear combination of its inputs (output activations from the preceding layer), with one of the weights acting as a trainable bias connected to a constant input. The transformation, called activation function, is a bounded non-decreasing (non-linear) function.

See Also:

Serialized Form

Field Summary

Fields
Modifier and Type	Field and Description
`protected double`	`epsilon` A small constant for numerical stability in RMSProp.
`protected double`	`lambda` The L2 regularization factor, which is also the weight decay factor.
`protected smile.math.TimeFunction`	`learningRate` The learning rate.
`protected smile.math.TimeFunction`	`momentum` The momentum factor.
`protected Layer[]`	`net` The hidden layers.
`protected OutputLayer`	`output` The output layer.
`protected int`	`p` The dimensionality of input data.
`protected double`	`rho` The discounting factor for the history/coming gradient in RMSProp.
`protected int`	`t` The training iterations.
`protected java.lang.ThreadLocal<double[]>`	`target` The buffer to store desired target value of training instance.

Constructor Summary

Constructors
Constructor and Description

MultilayerPerceptron(Layer... net)
Constructor.

Constructors
Constructor and Description
`MultilayerPerceptron(Layer... net)` Constructor.

Method Summary

All Methods Instance Methods Concrete Methods
Modifier and Type	Method and Description
`protected void`	`backpropagate(double[] x, boolean update)` Propagates the errors back through the network.
`double`	`getLearningRate()` Returns the learning rate.
`double`	`getMomentum()` Returns the momentum factor.
`double`	`getWeightDecay()` Returns the weight decay factor.
`protected void`	`propagate(double[] x)` Propagates the signals through the neural network.
`void`	`setLearningRate(smile.math.TimeFunction rate)` Sets the learning rate.
`void`	`setMomentum(smile.math.TimeFunction momentum)` Sets the momentum factor.
`void`	`setRMSProp(double rho, double epsilon)` Sets RMSProp parameters.
`void`	`setWeightDecay(double lambda)` Sets the weight decay factor.
`java.lang.String`	`toString()`
`protected void`	`update(int m)` Updates the weights for mini-batch training.

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

- Field Detail
  - p
```
protected int p
```
    The dimensionality of input data.
  - output
```
protected OutputLayer output
```
    The output layer.
  - net
```
protected Layer[] net
```
    The hidden layers.
  - target
```
protected transient java.lang.ThreadLocal<double[]> target
```
    The buffer to store desired target value of training instance.
  - learningRate
```
protected smile.math.TimeFunction learningRate
```
    The learning rate.
  - momentum
```
protected smile.math.TimeFunction momentum
```
    The momentum factor.
  - rho
```
protected double rho
```
    The discounting factor for the history/coming gradient in RMSProp.
  - epsilon
```
protected double epsilon
```
    A small constant for numerical stability in RMSProp.
  - lambda
```
protected double lambda
```
    The L2 regularization factor, which is also the weight decay factor.
  - t
```
protected int t
```
    The training iterations.
- Constructor Detail
  - MultilayerPerceptron
```
public MultilayerPerceptron(Layer... net)
```
    Constructor.
    
    Parameters:
    
    net - the layers from bottom to top. The input layer should not be included.
- Method Detail
  - toString
```
public java.lang.String toString()
```
    Overrides:
    
    toString in class java.lang.Object
  - setLearningRate
```
public void setLearningRate(smile.math.TimeFunction rate)
```
    Sets the learning rate.
    
    Parameters:
    
    rate - the learning rate.
  - setMomentum
```
public void setMomentum(smile.math.TimeFunction momentum)
```
    Sets the momentum factor. momentum = 0.0 means no momentum.
    
    Parameters:
    
    momentum - the momentum factor.
  - setRMSProp
```
public void setRMSProp(double rho,
                       double epsilon)
```
    Sets RMSProp parameters.
    
    Parameters:
    
    rho - The discounting factor for the history/coming gradient.
    
    epsilon - A small constant for numerical stability.
  - setWeightDecay
```
public void setWeightDecay(double lambda)
```
    Sets the weight decay factor. After each weight update, every weight is simply "decayed" or shrunk according to w = w * (1 - 2 * eta * lambda).
  - getLearningRate
```
public double getLearningRate()
```
    Returns the learning rate.
  - getMomentum
```
public double getMomentum()
```
    Returns the momentum factor.
  - getWeightDecay
```
public double getWeightDecay()
```
    Returns the weight decay factor.
  - propagate
```
protected void propagate(double[] x)
```
    Propagates the signals through the neural network.
  - backpropagate
```
protected void backpropagate(double[] x,
                             boolean update)
```
    Propagates the errors back through the network.
    
    Parameters:
    
    update - the flag if update the weights directly. It should be false for (mini-)batch.
  - update
```
protected void update(int m)
```
    Updates the weights for mini-batch training.
    
    Parameters:
    
    m - the mini-batch size.

Class MultilayerPerceptron

Field Summary

Constructor Summary

Method Summary

Methods inherited from class java.lang.Object

Field Detail

p

output

net

target

learningRate

momentum

rho

epsilon

lambda

t

Constructor Detail

MultilayerPerceptron

Method Detail

toString

setLearningRate

setMomentum

setRMSProp

setWeightDecay

getLearningRate

getMomentum

getWeightDecay

propagate

backpropagate

update