Serialized Form (Deep Netts Core Pro 3.0.0 API)

Package deepnetts.data.norm
- Class deepnetts.data.norm.AbstractScaler extends Object implements Serializable
- Class deepnetts.data.norm.DecimalScaler extends AbstractScaler implements Serializable
  - Serialized Fields
    - inputDivisor
```
float[] inputDivisor
```
    - outputDivisor
```
float[] outputDivisor
```
- Class deepnetts.data.norm.MaxScaler extends AbstractScaler implements Serializable
  - Serialized Fields
    - maxInputs
```
Tensor maxInputs
```
    - maxOutputs
```
Tensor maxOutputs
```
- Class deepnetts.data.norm.MinMaxScaler extends AbstractScaler implements Serializable
  - Serialized Fields
    - minInput
```
Tensor minInput
```
    - maxInput
```
Tensor maxInput
```
    - minOutput
```
Tensor minOutput
```
    - maxOutput
```
Tensor maxOutput
```
    - inDivider
```
Tensor inDivider
```
    - outDivider
```
Tensor outDivider
```
- Class deepnetts.data.norm.RangeScaler extends AbstractScaler implements Serializable
  - Serialized Fields
    - min
```
float min
```
    - max
```
float max
```
    - range
```
float range
```
- Class deepnetts.data.norm.Standardizer extends AbstractScaler implements Serializable
  - Serialized Fields
    - mean
```
Tensor mean
```
    - std
```
Tensor std
```
Package deepnetts.net
- Class deepnetts.net.ConvolutionalNetwork extends NeuralNetwork<BackpropagationTrainer> implements Serializable
  
  serialVersionUID:
  
  7311052836990578126L
- Class deepnetts.net.FeedForwardNetwork extends NeuralNetwork<BackpropagationTrainer> implements Serializable
  
  serialVersionUID:
  
  5819940381359274290L
  - Serialized Fields
    - inputTensor
```
Tensor inputTensor
```
- Class deepnetts.net.NeuralNetwork extends Object implements Serializable
  
  serialVersionUID:
  
  1L
  - Serialization Methods
    - readObject
```
private void readObject(ObjectInputStream ois)
                 throws ClassNotFoundException,
                        IOException
```
      Performs additional initialization after loading the network.
      
      Throws:
      
      ClassNotFoundException
      
      IOException
  - Serialized Fields
    - trainer
```
Trainer trainer
```
    - layers
```
List<E> layers
```
      Collection of all layers in this network (including input(first), output(last) and hidden(in between)). As a minimum neural network must have an input and output layer.
    - lossFunction
```
LossFunction lossFunction
```
      Loss function Loss function represents total network error for some data, and network learns by minimizing that error. Commonly used types of loss functions are Mean Squared Error for regression problems and and Cross Entropy for classification problems.
    - inputLayer
```
InputLayer inputLayer
```
      Input layer. This layer accepts external inputs and sends them to the next layer
    - outputLayer
```
OutputLayer outputLayer
```
      Output layer. This layer is the final step of processing network's input and its output is network's output.
    - outputLabels
```
String[] outputLabels
```
      Labels for network outputs (classes)
    - inputWrapper
```
Tensor inputWrapper
```
    - preprocessing
```
Preprocessing<T> preprocessing
```
    - normalizer
```
AbstractScaler normalizer
```
    - label
```
String label
```
      Network's label
    - regularizationSum
```
float regularizationSum
```
Package deepnetts.net.layers
- Class deepnetts.net.layers.AbstractLayer extends Object implements Serializable
  
  serialVersionUID:
  
  -3972836675081087082L
  - Serialized Fields
    - prevLayer
```
AbstractLayer prevLayer
```
      Previous layer in network.
    - nextLayer
```
AbstractLayer nextLayer
```
      Next layer in network.
    - weights
```
Tensor weights
```
      Input weight matrix / connectivity matrix for previous layer. Used in FullyConnected and OutputLayer. MaxPooling does not have Weights and ConvolutionalLayer has weights in filters.
    - inputs
```
Tensor inputs
```
      Inputs to this layer. A reference to outputs in previous layer, or external input in input layer).
    - outputs
```
Tensor outputs
```
      Layer outputs.
    - deltas
```
Tensor deltas
```
      Deltas used for learning.
    - deltaWeights
```
Tensor deltaWeights
```
      Weight changes for current and previous iteration.
    - prevDeltaWeights
```
Tensor prevDeltaWeights
```
      Weight changes for current and previous iteration.
    - gradients
```
Tensor gradients
```
      Gradients of a loss function calculates during a backward pass.
    - activation
```
ActivationFunction activation
```
      Activation function for this layer.
    - activationType
```
ActivationType activationType
```
      Type of activation function for this layer.
    - learningRate
```
float learningRate
```
      Learning rate for this layer.
    - momentum
```
float momentum
```
    - regL2
```
float regL2
```
    - regL1
```
float regL1
```
    - optimizerType
```
OptimizerType optimizerType
```
    - batchMode
```
boolean batchMode
```
    - batchSize
```
int batchSize
```
    - width
```
int width
```
    - height
```
int height
```
    - depth
```
int depth
```
    - biases
```
float[] biases
```
    - deltaBiases
```
float[] deltaBiases
```
    - prevDeltaBiases
```
float[] prevDeltaBiases
```
    - trainable
```
boolean trainable
```
    - optimizer
```
Optimizer optimizer
```
    - randomWeightsType
```
RandomWeightsType randomWeightsType
```
- Class deepnetts.net.layers.ConvolutionalLayer extends AbstractLayer implements Serializable
  
  serialVersionUID:
  
  -3972836675081087082L
  - Serialization Methods
    - readObject
```
private void readObject(ObjectInputStream ois)
                 throws ClassNotFoundException,
                        IOException
```
      Throws:
      
      ClassNotFoundException
      
      IOException
  - Serialized Fields
    - filters
```
Tensor filters
```
      Convolutional filters. Filters are stored as tensors.
    - filterWidth
```
int filterWidth
```
      Convolutional filter width (columns)
    - filterHeight
```
int filterHeight
```
      Filter height (rows)
    - filterDepth
```
int filterDepth
```
      Filter depth, corresponds to number of channels in previous layer.
    - filterGroups
```
int filterGroups
```
    - inGroupSize
```
int inGroupSize
```
    - outGroupSize
```
int outGroupSize
```
    - stride
```
int stride
```
      Convolution step, 1 by default. Number of steps convolutional filter is moved during convolution. Commonly used values 1, 2, rarely 3
    - padding
```
int padding
```
      Border padding filled with zeros (0, 1 or 2) Usually half of the filter size
    - fCenterX
```
int fCenterX
```
    - fCenterY
```
int fCenterY
```
    - maxIdx
```
int[][][][] maxIdx
```
    - multithreaded
```
boolean multithreaded
```
- Class deepnetts.net.layers.FlattenLayer extends AbstractLayer implements Serializable
- Class deepnetts.net.layers.FullyConnectedLayer extends AbstractLayer implements Serializable
  
  serialVersionUID:
  
  -8172689320815816926L
  - Serialized Fields
    - useDropout
```
boolean useDropout
```
    - dropout
```
float dropout
```
- Class deepnetts.net.layers.InputLayer extends AbstractLayer implements Serializable
  
  serialVersionUID:
  
  6519195289402298178L
- Class deepnetts.net.layers.MaxPoolingLayer extends AbstractLayer implements Serializable
  
  serialVersionUID:
  
  6257978737942468865L
  - Serialized Fields
    - filterWidth
```
int filterWidth
```
      Filter dimensions. Commonly used 2x2 with stride 2
    - filterHeight
```
int filterHeight
```
      Filter dimensions. Commonly used 2x2 with stride 2
    - stride
```
int stride
```
      Filter step. Commonly used 2
    - maxIdx
```
int[][][][] maxIdx
```
      Max activation idxs. Remember idx of max output for each filter position. [channel][row][col][2]
- Class deepnetts.net.layers.OutputLayer extends AbstractLayer implements Serializable
  
  serialVersionUID:
  
  4319240027730054207L
  - Serialized Fields
    - outputErrors
```
float[] outputErrors
```
    - labels
```
String[] labels
```
    - lossType
```
LossType lossType
```
    - singleOutInput
```
float singleOutInput
```
- Class deepnetts.net.layers.SoftmaxOutputLayer extends OutputLayer implements Serializable
  
  serialVersionUID:
  
  609777460047517229L
Package deepnetts.net.layers.activation
- Class deepnetts.net.layers.activation.LeakyRelu extends Object implements Serializable
  - Serialized Fields
    - a
```
float a
```
- Class deepnetts.net.layers.activation.Linear extends Object implements Serializable
- Class deepnetts.net.layers.activation.Relu extends Object implements Serializable
- Class deepnetts.net.layers.activation.Sigmoid extends Object implements Serializable
- Class deepnetts.net.layers.activation.Tanh extends Object implements Serializable
Package deepnetts.net.loss
- Class deepnetts.net.loss.BinaryCrossEntropyLoss extends Object implements Serializable
  
  serialVersionUID:
  
  3236227875327534698L
  - Serialized Fields
    - outputError
```
float[] outputError
```
      Cross entropy derivative is dE/dy = (y-t) / y*(1-y) Since denominator is same as sigmoid derivative, they are canceled when calculating delta in output layer: delta = dE/dy * dy/ds = y - t See http://neuralnetworksanddeeplearning.com/chap3.html#introducing_the_cross-entropy_cost_function http://peterroelants.github.io/posts/neural_network_implementation_intermezzo01/
    - totalError
```
float totalError
```
    - patternCount
```
int patternCount
```
    - regularizationSum
```
float regularizationSum
```
    - outputLayer
```
OutputLayer outputLayer
```
- Class deepnetts.net.loss.CrossEntropyLoss extends Object implements Serializable
  
  serialVersionUID:
  
  3646652383636911829L
  - Serialized Fields
    - outputError
```
float[] outputError
```
      Sum over all outputs and training samples Bishop, pg. 245, eq. 6.185 dE/ds = (y - t) (when Softmax activation is used) http://peterroelants.github.io/posts/neural_network_implementation_intermezzo02/
    - targetIdx
```
int targetIdx
```
    - totalError
```
float totalError
```
    - patternCount
```
int patternCount
```
    - regularizationSum
```
float regularizationSum
```
- Class deepnetts.net.loss.MeanSquaredErrorLoss extends Object implements Serializable
  
  serialVersionUID:
  
  5004545721099801809L
  - Serialized Fields
    - outputError
```
float[] outputError
```
    - totalError
```
float totalError
```
    - patternCount
```
int patternCount
```
    - regularizationSum
```
float regularizationSum
```
Package deepnetts.net.train
- Class deepnetts.net.train.BackpropagationTrainer extends Object implements Serializable
  - Serialization Methods
    - readObject
```
private void readObject(ObjectInputStream ois)
                 throws ClassNotFoundException,
                        IOException
```
      This method needs to be overridden to initialize transient fields after deserialization.
      
      Throws:
      
      ClassNotFoundException
      
      IOException
  - Serialized Fields
    - maxEpochs
```
long maxEpochs
```
      Training stops when this number of epochs is reached regardless of the total network error. One epoch represents one pass of the entire training set.
    - stopError
```
float stopError
```
      Training stops once total training error has reached this value.
    - learningRate
```
float learningRate
```
      Global learning rate, which controls the step size for the weights adjustment. Effectively a percent of error used to change weight.
    - dropout
```
float dropout
```
    - optType
```
OptimizerType optType
```
      Optimization algorithm type
    - momentum
```
float momentum
```
      Global momentum parameter
    - batchMode
```
boolean batchMode
```
      Set to true to use batch mode training
    - batchSize
```
int batchSize
```
      Size of mini batch. When full batch is used, this equals training set size
    - stopTraining
```
boolean stopTraining
```
      Flag to stop training
    - epoch
```
int epoch
```
      Current training epoch
    - valLoss
```
float valLoss
```
      Value of loss function calculated on validation set
    - trainAccuracy
```
float trainAccuracy
```
    - valAccuracy
```
float valAccuracy
```
    - stopAccuracy
```
float stopAccuracy
```
    - totalTrainingLoss
```
float totalTrainingLoss
```
    - shuffle
```
boolean shuffle
```
      Shuffle training set before each epoch during training.
    - neuralNet
```
NeuralNetwork<T extends Trainer> neuralNet
```
      A neural network to train.
    - lossFunction
```
LossFunction lossFunction
```
    - trainingSnapshots
```
boolean trainingSnapshots
```
    - snapshotEpochs
```
int snapshotEpochs
```
    - snapshotPath
```
String snapshotPath
```
    - earlyStopping
```
boolean earlyStopping
```
      Use early stopping setting.
    - checkpointEpochs
```
int checkpointEpochs
```
      How many epochs for early stopping checkpoint.
    - earlyStoppingMinLossChange
```
float earlyStoppingMinLossChange
```
      Min delta between checkpoints to continue training
    - earlyStoppingPatience
```
int earlyStoppingPatience
```
      How many checkpoints to wait before stopping training
    - earlyStoppingCheckpointCount
```
int earlyStoppingCheckpointCount
```
    - prevCheckpointLoss
```
float prevCheckpointLoss
```
    - regL2
```
float regL2
```
    - regL1
```
float regL1
```
    - lossHistory
```
LinkedList<E> lossHistory
```
    - avgConvergenceSpeeed
```
float[] avgConvergenceSpeeed
```
    - extendedLogging
```
boolean extendedLogging
```
Package deepnetts.net.train.opt
- Class deepnetts.net.train.opt.AdaGradOptimizer extends Object implements Serializable
  
  serialVersionUID:
  
  2833280391108408107L
  - Serialized Fields
    - learningRate
```
float learningRate
```
    - prevGradSqrSum
```
Tensor prevGradSqrSum
```
    - prevBiasSqrSum
```
Tensor prevBiasSqrSum
```
- Class deepnetts.net.train.opt.MomentumOptimizer extends Object implements Serializable
  
  serialVersionUID:
  
  6936741415174730939L
  - Serialized Fields
    - momentum
```
float momentum
```
    - learningRate
```
float learningRate
```
    - prevDeltaWeights
```
Tensor prevDeltaWeights
```
    - prevDeltaBiases
```
float[] prevDeltaBiases
```
- Class deepnetts.net.train.opt.RmsPropOptimizer extends Object implements Serializable
  
  serialVersionUID:
  
  2804113940324244785L
  - Serialized Fields
    - learningRate
```
float learningRate
```
    - movAvgGradSqrSum
```
Tensor movAvgGradSqrSum
```
    - movAvgBiasSqrSum
```
Tensor movAvgBiasSqrSum
```
    - beta
```
float beta
```
- Class deepnetts.net.train.opt.SgdOptimizer extends Object implements Serializable
  
  serialVersionUID:
  
  7408169634780483571L
  - Serialized Fields
    - learningRate
```
float learningRate
```
Package deepnetts.util
- Class deepnetts.util.DeepNettsException extends RuntimeException implements Serializable
- Class deepnetts.util.ImagePreprocessing extends Object implements Serializable
  - Serialized Fields
    - subMean
```
boolean subMean
```
    - invertPixels
```
boolean invertPixels
```
    - scaleImage
```
boolean scaleImage
```
    - mean
```
Tensor mean
```
    - isEnabled
```
boolean isEnabled
```
- Class deepnetts.util.LicenseChecker.LicenceException extends RuntimeException implements Serializable
- Class deepnetts.util.Tensor extends Object implements Serializable
  
  serialVersionUID:
  
  8123299065043860349L
  - Serialized Fields
    - cols
```
int cols
```
    - rows
```
int rows
```
    - depth
```
int depth
```
    - fourthDim
```
int fourthDim
```
    - rank
```
int rank
```
    - shape
```
int[] shape
```
    - values
```
float[] values
```
      Values stored in this tensor make it final , only input layer and tests sets values
- Class deepnetts.util.TypedProperties extends Properties implements Serializable

Serialized Form

Package deepnetts.data.norm

Class deepnetts.data.norm.AbstractScaler extends Object implements Serializable

Class deepnetts.data.norm.DecimalScaler extends AbstractScaler implements Serializable

Serialized Fields

inputDivisor

outputDivisor

Class deepnetts.data.norm.MaxScaler extends AbstractScaler implements Serializable

Serialized Fields

maxInputs

maxOutputs

Class deepnetts.data.norm.MinMaxScaler extends AbstractScaler implements Serializable

Serialized Fields

minInput

maxInput

minOutput

maxOutput

inDivider

outDivider

Class deepnetts.data.norm.RangeScaler extends AbstractScaler implements Serializable

Serialized Fields

min

max

range

Class deepnetts.data.norm.Standardizer extends AbstractScaler implements Serializable

Serialized Fields

mean

std

Package deepnetts.net

Class deepnetts.net.ConvolutionalNetwork extends NeuralNetwork<BackpropagationTrainer> implements Serializable

Class deepnetts.net.FeedForwardNetwork extends NeuralNetwork<BackpropagationTrainer> implements Serializable

Serialized Fields

inputTensor

Class deepnetts.net.NeuralNetwork extends Object implements Serializable

Serialization Methods

readObject

Serialized Fields

trainer

layers

lossFunction

inputLayer

outputLayer

outputLabels

inputWrapper

preprocessing

normalizer

label

regularizationSum

Package deepnetts.net.layers

Class deepnetts.net.layers.AbstractLayer extends Object implements Serializable

Serialized Fields

prevLayer

nextLayer

weights

inputs

outputs

deltas

deltaWeights

prevDeltaWeights

gradients

activation

activationType

learningRate

momentum

regL2

regL1

optimizerType

batchMode

batchSize

width

height

depth

biases

deltaBiases

prevDeltaBiases

trainable

optimizer

randomWeightsType

Class deepnetts.net.layers.ConvolutionalLayer extends AbstractLayer implements Serializable

Serialization Methods