org.deeplearning4j.nn.modelimport.keras.layers

Class KerasConvolution

java.lang.Object

org.deeplearning4j.nn.modelimport.keras.KerasLayer

org.deeplearning4j.nn.modelimport.keras.layers.KerasConvolution

public class KerasConvolution
extends KerasLayer

Imports a Convolution layer from Keras.

Nested Class Summary

Nested classes/interfaces inherited from class org.deeplearning4j.nn.modelimport.keras.KerasLayer

KerasLayer.DimOrder

Field Summary

Fields

Modifier and Type	Field and Description
static java.lang.String	KERAS_PARAM_NAME_B
static java.lang.String	KERAS_PARAM_NAME_W
static int	NUM_TRAINABLE_PARAMS

Fields inherited from class org.deeplearning4j.nn.modelimport.keras.KerasLayer

className, DIM_ORDERING_TENSORFLOW, DIM_ORDERING_THEANO, dimOrder, dropout, inboundLayerNames, INIT_GLOROT_NORMAL, INIT_GLOROT_UNIFORM, INIT_HE_NORMAL, INIT_HE_UNIFORM, INIT_IDENTITY, INIT_LECUN_UNIFORM, INIT_NORMAL, INIT_ORTHOGONAL, INIT_UNIFORM, INIT_ZERO, inputShape, KERAS_ACTIVATION_HARD_SIGMOID, KERAS_ACTIVATION_LINEAR, KERAS_ACTIVATION_RELU, KERAS_ACTIVATION_SIGMOID, KERAS_ACTIVATION_SOFTMAX, KERAS_ACTIVATION_SOFTPLUS, KERAS_ACTIVATION_SOFTSIGN, KERAS_ACTIVATION_TANH, KERAS_LOSS_BINARY_CROSSENTROPY, KERAS_LOSS_CATEGORICAL_CROSSENTROPY, KERAS_LOSS_COSINE_PROXIMITY, KERAS_LOSS_HINGE, KERAS_LOSS_KLD, KERAS_LOSS_KULLBACK_LEIBLER_DIVERGENCE, KERAS_LOSS_MAE, KERAS_LOSS_MAPE, KERAS_LOSS_MEAN_ABSOLUTE_ERROR, KERAS_LOSS_MEAN_ABSOLUTE_PERCENTAGE_ERROR, KERAS_LOSS_MEAN_SQUARED_ERROR, KERAS_LOSS_MEAN_SQUARED_LOGARITHMIC_ERROR, KERAS_LOSS_MSE, KERAS_LOSS_MSLE, KERAS_LOSS_POISSON, KERAS_LOSS_SPARSE_CATEGORICAL_CROSSENTROPY, KERAS_LOSS_SQUARED_HINGE, layer, LAYER_BORDER_MODE_FULL, LAYER_BORDER_MODE_SAME, LAYER_BORDER_MODE_VALID, LAYER_CLASS_NAME_ACTIVATION, LAYER_CLASS_NAME_AVERAGE_POOLING_1D, LAYER_CLASS_NAME_AVERAGE_POOLING_2D, LAYER_CLASS_NAME_BATCHNORMALIZATION, LAYER_CLASS_NAME_CONVOLUTION_1D, LAYER_CLASS_NAME_CONVOLUTION_2D, LAYER_CLASS_NAME_DENSE, LAYER_CLASS_NAME_DROPOUT, LAYER_CLASS_NAME_EMBEDDING, LAYER_CLASS_NAME_FLATTEN, LAYER_CLASS_NAME_GLOBAL_AVERAGE_POOLING_1D, LAYER_CLASS_NAME_GLOBAL_AVERAGE_POOLING_2D, LAYER_CLASS_NAME_GLOBAL_MAX_POOLING_1D, LAYER_CLASS_NAME_GLOBAL_MAX_POOLING_2D, LAYER_CLASS_NAME_INPUT, LAYER_CLASS_NAME_LSTM, LAYER_CLASS_NAME_MAX_POOLING_1D, LAYER_CLASS_NAME_MAX_POOLING_2D, LAYER_CLASS_NAME_MERGE, LAYER_CLASS_NAME_TIME_DISTRIBUTED, LAYER_CLASS_NAME_TIME_DISTRIBUTED_DENSE, LAYER_CLASS_NAME_ZERO_PADDING_1D, LAYER_CLASS_NAME_ZERO_PADDING_2D, LAYER_FIELD_ACTIVATION, LAYER_FIELD_B_REGULARIZER, LAYER_FIELD_BATCH_INPUT_SHAPE, LAYER_FIELD_BORDER_MODE, LAYER_FIELD_CLASS_NAME, LAYER_FIELD_CONFIG, LAYER_FIELD_DIM_ORDERING, LAYER_FIELD_DROPOUT, LAYER_FIELD_DROPOUT_W, LAYER_FIELD_INBOUND_NODES, LAYER_FIELD_INIT, LAYER_FIELD_LAYER, LAYER_FIELD_NAME, LAYER_FIELD_NB_COL, LAYER_FIELD_NB_FILTER, LAYER_FIELD_NB_ROW, LAYER_FIELD_OUTPUT_DIM, LAYER_FIELD_POOL_SIZE, LAYER_FIELD_STRIDES, LAYER_FIELD_SUBSAMPLE, LAYER_FIELD_W_REGULARIZER, layerName, REGULARIZATION_TYPE_L1, REGULARIZATION_TYPE_L2, vertex, weightL1Regularization, weightL2Regularization, weights

Constructor Summary

Constructors

Constructor and Description
KerasConvolution(java.util.Map<java.lang.String,java.lang.Object> layerConfig) Constructor from parsed Keras layer configuration dictionary.
KerasConvolution(java.util.Map<java.lang.String,java.lang.Object> layerConfig, boolean enforceTrainingConfig) Constructor from parsed Keras layer configuration dictionary.

Method Summary

Modifier and Type	Method and Description
ConvolutionLayer	getConvolutionLayer() Get DL4J ConvolutionLayer.
int	getNumParams() Returns number of trainable parameters in layer.
InputType	getOutputType(InputType... inputType) Get layer output type.
void	setWeights(java.util.Map<java.lang.String,org.nd4j.linalg.api.ndarray.INDArray> weights) Set weights for layer.

Methods inherited from class org.deeplearning4j.nn.modelimport.keras.KerasLayer

checkForUnsupportedConfigurations, copyWeightsToLayer, getActivationFromConfig, getBiasL1RegularizationFromConfig, getClassName, getClassNameFromConfig, getConvolutionModeFromConfig, getDimOrder, getDropoutFromConfig, getInboundLayerNames, getInboundLayerNamesFromConfig, getInnerLayerConfigFromConfig, getInputPreprocessor, getInputShape, getKerasLayerFromConfig, getKerasLayerFromConfig, getKernelSizeFromConfig, getLayer, getLayerName, getLayerNameFromConfig, getNOutFromConfig, getPaddingFromBorderModeConfig, getStrideFromConfig, getTimeDistributedLayerConfig, getVertex, getWeightInitFromConfig, getWeightL1RegularizationFromConfig, getWeightL2RegularizationFromConfig, isInputPreProcessor, isLayer, isValidInboundLayer, isVertex, mapActivation, mapLossFunction, mapPoolingDimensions, mapPoolingType, mapWeightInitialization, setDimOrder, setInboundLayerNames, usesRegularization

Methods inherited from class java.lang.Object

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

Field Detail

NUM_TRAINABLE_PARAMS

public static final int NUM_TRAINABLE_PARAMS

See Also:

Constant Field Values

KERAS_PARAM_NAME_W

public static final java.lang.String KERAS_PARAM_NAME_W

See Also:

Constant Field Values

KERAS_PARAM_NAME_B

public static final java.lang.String KERAS_PARAM_NAME_B

See Also:

Constant Field Values

Constructor Detail

KerasConvolution

public KerasConvolution(java.util.Map<java.lang.String,java.lang.Object> layerConfig)
                 throws InvalidKerasConfigurationException,
                        UnsupportedKerasConfigurationException

Constructor from parsed Keras layer configuration dictionary.

Parameters:

layerConfig - dictionary containing Keras layer configuration

Throws:

InvalidKerasConfigurationException

UnsupportedKerasConfigurationException

KerasConvolution

public KerasConvolution(java.util.Map<java.lang.String,java.lang.Object> layerConfig,
                        boolean enforceTrainingConfig)
                 throws InvalidKerasConfigurationException,
                        UnsupportedKerasConfigurationException

Constructor from parsed Keras layer configuration dictionary.

Parameters:

layerConfig - dictionary containing Keras layer configuration

enforceTrainingConfig - whether to enforce training-related configuration options

Throws:

InvalidKerasConfigurationException

UnsupportedKerasConfigurationException

Method Detail

getConvolutionLayer

public ConvolutionLayer getConvolutionLayer()

Get DL4J ConvolutionLayer.

Returns:

ConvolutionLayer

getOutputType

public InputType getOutputType(InputType... inputType)
                        throws InvalidKerasConfigurationException

Get layer output type.

Overrides:

getOutputType in class KerasLayer

Parameters:

inputType - Array of InputTypes

Returns:

output type as InputType

Throws:

InvalidKerasConfigurationException

getNumParams

public int getNumParams()

Returns number of trainable parameters in layer.

Overrides:

getNumParams in class KerasLayer

Returns:

number of trainable parameters (2)

setWeights

public void setWeights(java.util.Map<java.lang.String,org.nd4j.linalg.api.ndarray.INDArray> weights)
                throws InvalidKerasConfigurationException

Set weights for layer.

Overrides:

setWeights in class KerasLayer

Parameters:

weights - Map from parameter name to INDArray.

Throws:

InvalidKerasConfigurationException