org.deeplearning4j.nn.modelimport.keras

Class KerasSequentialModel

java.lang.Object

org.deeplearning4j.nn.modelimport.keras.KerasModel

org.deeplearning4j.nn.modelimport.keras.KerasSequentialModel

public class KerasSequentialModel
extends KerasModel

Build DL4J MultiLayerNetwork model from Keras Sequential model configuration.

Field Summary

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

className, enforceTrainingConfig, HDF5_MODEL_CONFIG_ATTRIBUTE, HDF5_MODEL_WEIGHTS_ROOT, HDF5_TRAINING_CONFIG_ATTRIBUTE, inputLayerNames, layers, layersOrdered, MODEL_CLASS_NAME_MODEL, MODEL_CLASS_NAME_SEQUENTIAL, MODEL_CONFIG_FIELD_INPUT_LAYERS, MODEL_CONFIG_FIELD_LAYERS, MODEL_CONFIG_FIELD_OUTPUT_LAYERS, MODEL_FIELD_CLASS_NAME, MODEL_FIELD_CONFIG, outputLayerNames, outputTypes, TRAINING_CONFIG_FIELD_LOSS, truncatedBPTT, useTruncatedBPTT

Constructor Summary

Constructors

Modifier	Constructor and Description
protected	KerasSequentialModel()
	KerasSequentialModel(org.deeplearning4j.nn.modelimport.keras.KerasModel.ModelBuilder modelBuilder) (Recommended) Builder-pattern constructor for Sequential model.
	KerasSequentialModel(java.lang.String modelJson, java.lang.String modelYaml, Hdf5Archive weightsArchive, java.lang.String weightsRoot, java.lang.String trainingJson, Hdf5Archive trainingArchive, boolean enforceTrainingConfig) (Not recommended) Constructor for Sequential model from model configuration (JSON or YAML), training configuration (JSON), weights, and "training mode" boolean indicator.

Method Summary

Modifier and Type	Method and Description
MultiLayerConfiguration	getMultiLayerConfiguration() Configure a MultiLayerConfiguration from this Keras Sequential model configuration.
MultiLayerNetwork	getMultiLayerNetwork() Build a MultiLayerNetwork from this Keras Sequential model configuration.
MultiLayerNetwork	getMultiLayerNetwork(boolean importWeights) Build a MultiLayerNetwork from this Keras Sequential model configuration and import weights.

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

getComputationGraph, getComputationGraph, getComputationGraphConfiguration, helperCopyWeightsToModel, helperImportTrainingConfiguration, helperImportWeights, helperInferOutputTypes, helperPrepareLayers, parseJsonString, parseYamlString

Methods inherited from class java.lang.Object

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

Constructor Detail

KerasSequentialModel

public KerasSequentialModel(org.deeplearning4j.nn.modelimport.keras.KerasModel.ModelBuilder modelBuilder)
                     throws UnsupportedKerasConfigurationException,
                            java.io.IOException,
                            InvalidKerasConfigurationException

(Recommended) Builder-pattern constructor for Sequential model.

Parameters:

modelBuilder - builder object

Throws:

java.io.IOException

InvalidKerasConfigurationException

UnsupportedKerasConfigurationException

KerasSequentialModel

public KerasSequentialModel(java.lang.String modelJson,
                            java.lang.String modelYaml,
                            Hdf5Archive weightsArchive,
                            java.lang.String weightsRoot,
                            java.lang.String trainingJson,
                            Hdf5Archive trainingArchive,
                            boolean enforceTrainingConfig)
                     throws java.io.IOException,
                            InvalidKerasConfigurationException,
                            UnsupportedKerasConfigurationException

(Not recommended) Constructor for Sequential model from model configuration (JSON or YAML), training configuration (JSON), weights, and "training mode" boolean indicator. When built in training mode, certain unsupported configurations (e.g., unknown regularizers) will throw Exceptions. When enforceTrainingConfig=false, these will generate warnings but will be otherwise ignored.

Parameters:

modelJson - model configuration JSON string

modelYaml - model configuration YAML string

trainingJson - training configuration JSON string

Throws:

java.io.IOException

InvalidKerasConfigurationException

UnsupportedKerasConfigurationException

KerasSequentialModel

protected KerasSequentialModel()

Method Detail

getMultiLayerConfiguration

public MultiLayerConfiguration getMultiLayerConfiguration()
                                                   throws InvalidKerasConfigurationException,
                                                          UnsupportedKerasConfigurationException

Configure a MultiLayerConfiguration from this Keras Sequential model configuration.

Returns:

MultiLayerConfiguration

Throws:

InvalidKerasConfigurationException

UnsupportedKerasConfigurationException

getMultiLayerNetwork

public MultiLayerNetwork getMultiLayerNetwork()
                                       throws InvalidKerasConfigurationException,
                                              UnsupportedKerasConfigurationException

Build a MultiLayerNetwork from this Keras Sequential model configuration.

Returns:

MultiLayerNetwork

Throws:

InvalidKerasConfigurationException

UnsupportedKerasConfigurationException

getMultiLayerNetwork

public MultiLayerNetwork getMultiLayerNetwork(boolean importWeights)
                                       throws InvalidKerasConfigurationException,
                                              UnsupportedKerasConfigurationException

Build a MultiLayerNetwork from this Keras Sequential model configuration and import weights.

Returns:

MultiLayerNetwork

Throws:

InvalidKerasConfigurationException

UnsupportedKerasConfigurationException