org.deeplearning4j.nn.modelimport.keras

Class Model

java.lang.Object

org.deeplearning4j.nn.modelimport.keras.Model

Deprecated.

Use KerasModelImport instead.

@Deprecated
public class Model
extends java.lang.Object

Routines for importing saved Keras models.

Method Summary

Modifier and Type	Method and Description
static ComputationGraph	importFunctionalApiModel(java.lang.String modelHdf5Filename) Deprecated. Use KerasModelImport.importKerasModelAndWeights(java.io.InputStream, boolean) instead
static ComputationGraph	importFunctionalApiModel(java.lang.String configJsonFilename, java.lang.String weightsHdf5Filename) Deprecated. Use KerasModelImport.importKerasModelAndWeights(java.io.InputStream, boolean) instead
static ComputationGraph	importFunctionalApiModelInputStream(java.io.InputStream modelHdf5Stream) Deprecated. Use KerasModelImport.importKerasModelAndWeights(java.io.InputStream, boolean) instead
static Model	importModel(java.lang.String modelHdf5Filename) Deprecated. Use KerasModelImport.importKerasModelAndWeights(java.io.InputStream, boolean) instead
static Model	importModel(java.lang.String configJsonFilename, java.lang.String weightsHdf5Filename) Deprecated. Use KerasModelImport.importKerasModelAndWeights(java.io.InputStream, boolean) instead
static Model	importModelInputStream(java.io.InputStream modelHdf5Stream) Deprecated. Use KerasModelImport.importKerasModelAndWeights(java.io.InputStream, boolean) instead
static MultiLayerNetwork	importSequentialModel(java.lang.String modelHdf5Filename) Deprecated. Use KerasModelImport.importKerasSequentialModelAndWeights(java.io.InputStream, boolean) instead
static MultiLayerNetwork	importSequentialModel(java.lang.String configJsonFilename, java.lang.String weightsHdf5Filename) Deprecated. Use KerasModelImport.importKerasSequentialModelAndWeights(java.io.InputStream, boolean) instead
static MultiLayerNetwork	importSequentialModelInputStream(java.io.InputStream modelHdf5Stream) Deprecated. Use KerasModelImport.importKerasSequentialModelAndWeights(java.io.InputStream, boolean) instead

Methods inherited from class java.lang.Object

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

Method Detail

importModelInputStream

@Deprecated
public static Model importModelInputStream(java.io.InputStream modelHdf5Stream)
                                                throws UnsupportedKerasConfigurationException,
                                                       java.io.IOException,
                                                       InvalidKerasConfigurationException

Deprecated. Use KerasModelImport.importKerasModelAndWeights(java.io.InputStream, boolean) instead

Load Keras model saved using model.save_model(...).

Parameters:

modelHdf5Stream - input stream storing Keras Sequential model

Returns:

DL4J MultiLayerNetwork

Throws:

UnsupportedKerasConfigurationException

java.io.IOException

Model

InvalidKerasConfigurationException

See Also:

MultiLayerNetwork

importSequentialModelInputStream

@Deprecated
public static MultiLayerNetwork importSequentialModelInputStream(java.io.InputStream modelHdf5Stream)
                                                                      throws UnsupportedKerasConfigurationException,
                                                                             java.io.IOException,
                                                                             InvalidKerasConfigurationException

Deprecated. Use KerasModelImport.importKerasSequentialModelAndWeights(java.io.InputStream, boolean) instead

Imports a Keras Sequential model saved using model.save_model(...). Model configuration and weights are loaded from single HDF5 archive.

Parameters:

modelHdf5Stream - input stream storing Keras Sequential model

Returns:

DL4J MultiLayerNetwork

Throws:

UnsupportedKerasConfigurationException

java.io.IOException

InvalidKerasConfigurationException

See Also:

MultiLayerNetwork

importFunctionalApiModelInputStream

@Deprecated
public static ComputationGraph importFunctionalApiModelInputStream(java.io.InputStream modelHdf5Stream)
                                                                        throws UnsupportedKerasConfigurationException,
                                                                               java.io.IOException,
                                                                               InvalidKerasConfigurationException

Deprecated. Use KerasModelImport.importKerasModelAndWeights(java.io.InputStream, boolean) instead

Imports a Keras Functional API model saved using model.save_model(...). Model configuration and weights are loaded from single HDF5 archive.

Parameters:

modelHdf5Stream - input stream storing storing Keras Functional API model

Returns:

DL4J ComputationGraph

Throws:

UnsupportedKerasConfigurationException

java.io.IOException

InvalidKerasConfigurationException

importModel

@Deprecated
public static Model importModel(java.lang.String modelHdf5Filename)
                                     throws java.io.IOException,
                                            InvalidKerasConfigurationException,
                                            UnsupportedKerasConfigurationException

Deprecated. Use KerasModelImport.importKerasModelAndWeights(java.io.InputStream, boolean) instead

Load Keras model saved using model.save_model(...).

Parameters:

modelHdf5Filename - path to HDF5 archive storing Keras model

Returns:

DL4J Model interface

Throws:

java.io.IOException

InvalidKerasConfigurationException

UnsupportedKerasConfigurationException

See Also:

Model

importModel

@Deprecated
public static Model importModel(java.lang.String configJsonFilename,
                                            java.lang.String weightsHdf5Filename)
                                     throws java.io.IOException,
                                            InvalidKerasConfigurationException,
                                            UnsupportedKerasConfigurationException

Deprecated. Use KerasModelImport.importKerasModelAndWeights(java.io.InputStream, boolean) instead

Load Keras model where the config and weights were saved separately using calls to model.to_json() and model.save_weights(...).

Parameters:

configJsonFilename - path to JSON file storing Keras Functional API model configuration

weightsHdf5Filename - path to HDF5 archive storing Keras Functional API model weights

Returns:

DL4J Model interface

Throws:

java.io.IOException

InvalidKerasConfigurationException

UnsupportedKerasConfigurationException

See Also:

Model

importSequentialModel

@Deprecated
public static MultiLayerNetwork importSequentialModel(java.lang.String modelHdf5Filename)
                                                           throws java.io.IOException,
                                                                  InvalidKerasConfigurationException,
                                                                  UnsupportedKerasConfigurationException

Deprecated. Use KerasModelImport.importKerasSequentialModelAndWeights(java.io.InputStream, boolean) instead

Imports a Keras Sequential model saved using model.save_model(...). Model configuration and weights are loaded from single HDF5 archive.

Parameters:

modelHdf5Filename - path to HDF5 archive storing Keras Sequential model

Returns:

DL4J MultiLayerNetwork

Throws:

java.io.IOException

InvalidKerasConfigurationException

UnsupportedKerasConfigurationException

See Also:

MultiLayerNetwork

importFunctionalApiModel

@Deprecated
public static ComputationGraph importFunctionalApiModel(java.lang.String modelHdf5Filename)
                                                             throws java.io.IOException,
                                                                    InvalidKerasConfigurationException,
                                                                    UnsupportedKerasConfigurationException

Deprecated. Use KerasModelImport.importKerasModelAndWeights(java.io.InputStream, boolean) instead

Imports a Keras Functional API model saved using model.save_model(...). Model configuration and weights are loaded from single HDF5 archive.

Parameters:

modelHdf5Filename - path to HDF5 archive storing Keras Functional API model

Returns:

DL4J ComputationGraph

Throws:

java.io.IOException

InvalidKerasConfigurationException

UnsupportedKerasConfigurationException

See Also:

ComputationGraph

importSequentialModel

@Deprecated
public static MultiLayerNetwork importSequentialModel(java.lang.String configJsonFilename,
                                                                  java.lang.String weightsHdf5Filename)
                                                           throws java.io.IOException,
                                                                  InvalidKerasConfigurationException,
                                                                  UnsupportedKerasConfigurationException

Deprecated. Use KerasModelImport.importKerasSequentialModelAndWeights(java.io.InputStream, boolean) instead

Imports a Keras Sequential model saved using model.save_model(...). Model configuration and weights are loaded from single HDF5 archive.

Parameters:

configJsonFilename - path to JSON file storing Keras Functional API model configuration

weightsHdf5Filename - path to HDF5 archive storing Keras Functional API model weights

Returns:

DL4J MultiLayerNetwork

Throws:

java.io.IOException

InvalidKerasConfigurationException

UnsupportedKerasConfigurationException

See Also:

MultiLayerNetwork

importFunctionalApiModel

@Deprecated
public static ComputationGraph importFunctionalApiModel(java.lang.String configJsonFilename,
                                                                    java.lang.String weightsHdf5Filename)
                                                             throws java.io.IOException,
                                                                    InvalidKerasConfigurationException,
                                                                    UnsupportedKerasConfigurationException

Deprecated. Use KerasModelImport.importKerasModelAndWeights(java.io.InputStream, boolean) instead

Imports a Keras Functional API model saved using model.save_model(...). Model configuration and weights are loaded from single HDF5 archive.

Parameters:

configJsonFilename - path to JSON file storing Keras Functional API model configuration

weightsHdf5Filename - path to HDF5 archive storing Keras Functional API model weights

Returns:

DL4J ComputationGraph

Throws:

java.io.IOException

InvalidKerasConfigurationException

UnsupportedKerasConfigurationException

See Also:

ComputationGraph