Package org.deeplearning4j.nn.conf
Class MultiLayerConfiguration
- java.lang.Object
-
- org.deeplearning4j.nn.conf.MultiLayerConfiguration
-
- All Implemented Interfaces:
Serializable
,Cloneable
public class MultiLayerConfiguration extends Object implements Serializable, Cloneable
- See Also:
- Serialized Form
-
-
Nested Class Summary
Nested Classes Modifier and Type Class Description static class
MultiLayerConfiguration.Builder
-
Field Summary
Fields Modifier and Type Field Description protected BackpropType
backpropType
protected CacheMode
cacheMode
protected List<NeuralNetConfiguration>
confs
protected DataType
dataType
protected int
epochCount
protected WorkspaceMode
inferenceWorkspaceMode
protected Map<Integer,InputPreProcessor>
inputPreProcessors
protected int
iterationCount
protected int
tbpttBackLength
protected int
tbpttFwdLength
protected WorkspaceMode
trainingWorkspaceMode
protected boolean
validateOutputLayerConfig
-
Constructor Summary
Constructors Constructor Description MultiLayerConfiguration()
-
Method Summary
All Methods Static Methods Instance Methods Concrete Methods Modifier and Type Method Description MultiLayerConfiguration
clone()
static MultiLayerConfiguration
fromJson(String json)
Create a neural net configuration from jsonstatic MultiLayerConfiguration
fromYaml(String json)
Create a neural net configuration from jsonNeuralNetConfiguration
getConf(int i)
int
getEpochCount()
InputPreProcessor
getInputPreProcess(int curr)
List<InputType>
getLayerActivationTypes(@NonNull InputType inputType)
For the given input shape/type for the network, return a list of activation sizes for each layer in the network.
i.e., list.get(i) is the output activation sizes for layer iNetworkMemoryReport
getMemoryReport(InputType inputType)
Get aMemoryReport
for the given MultiLayerConfiguration.void
setEpochCount(int epochCount)
String
toJson()
String
toString()
String
toYaml()
-
-
-
Field Detail
-
confs
protected List<NeuralNetConfiguration> confs
-
inputPreProcessors
protected Map<Integer,InputPreProcessor> inputPreProcessors
-
backpropType
protected BackpropType backpropType
-
tbpttFwdLength
protected int tbpttFwdLength
-
tbpttBackLength
protected int tbpttBackLength
-
validateOutputLayerConfig
protected boolean validateOutputLayerConfig
-
trainingWorkspaceMode
protected WorkspaceMode trainingWorkspaceMode
-
inferenceWorkspaceMode
protected WorkspaceMode inferenceWorkspaceMode
-
cacheMode
protected CacheMode cacheMode
-
dataType
protected DataType dataType
-
iterationCount
protected int iterationCount
-
epochCount
protected int epochCount
-
-
Method Detail
-
getEpochCount
public int getEpochCount()
-
setEpochCount
public void setEpochCount(int epochCount)
-
toYaml
public String toYaml()
- Returns:
- JSON representation of NN configuration
-
fromYaml
public static MultiLayerConfiguration fromYaml(String json)
Create a neural net configuration from json- Parameters:
json
- the neural net configuration from json- Returns:
MultiLayerConfiguration
-
toJson
public String toJson()
- Returns:
- JSON representation of NN configuration
-
fromJson
public static MultiLayerConfiguration fromJson(String json)
Create a neural net configuration from json- Parameters:
json
- the neural net configuration from json- Returns:
MultiLayerConfiguration
-
getConf
public NeuralNetConfiguration getConf(int i)
-
clone
public MultiLayerConfiguration clone()
-
getInputPreProcess
public InputPreProcessor getInputPreProcess(int curr)
-
getMemoryReport
public NetworkMemoryReport getMemoryReport(InputType inputType)
Get aMemoryReport
for the given MultiLayerConfiguration. This is used to estimate the memory requirements for the given network configuration and input- Parameters:
inputType
- Input types for the network- Returns:
- Memory report for the network
-
getLayerActivationTypes
public List<InputType> getLayerActivationTypes(@NonNull @NonNull InputType inputType)
For the given input shape/type for the network, return a list of activation sizes for each layer in the network.
i.e., list.get(i) is the output activation sizes for layer i- Parameters:
inputType
- Input type for the network- Returns:
- A lits of activation types for the network, indexed by layer number
-
-