Package org.ojalgo.ann
Class WrappedANN
java.lang.Object
org.ojalgo.ann.WrappedANN
- All Implemented Interfaces:
Supplier<ArtificialNeuralNetwork>
- Direct Known Subclasses:
NetworkInvoker
,NetworkTrainer
-
Field Summary
FieldsModifier and TypeFieldDescriptionprivate final int
private PhysicalStore
<Double> private final ArtificialNeuralNetwork
private final PhysicalStore<Double>[]
-
Constructor Summary
Constructors -
Method Summary
Modifier and TypeMethodDescription(package private) void
adjust
(int layer, PhysicalStore<Double> input, PhysicalStore<Double> output, PhysicalStore<Double> upstreamGradient, PhysicalStore<Double> downstreamGradient) (package private) int
depth()
boolean
get()
(package private) ArtificialNeuralNetwork.Activator
getActivator
(int layer) (package private) int
(package private) double
getBias
(int layer, int output) (package private) PhysicalStore
<Double> getInput()
(package private) PhysicalStore
<Double> getInput
(int layer) (package private) PhysicalStore
<Double> (package private) PhysicalStore
<Double> getOutput
(int layer) (package private) ArtificialNeuralNetwork.Activator
(package private) double
getWeight
(int layer, int input, int output) (package private) List
<MatrixStore<Double>> int
hashCode()
(package private) MatrixStore
<Double> invoke
(Access1D<Double> input, TrainingConfiguration configuration) When usingNetworkTrainer
orNetworkInvoker
with a batch size larger than 1 this utility may help with creating the batches.(package private) DataBatch
(package private) void
(package private) void
setActivator
(int layer, ArtificialNeuralNetwork.Activator activator) (package private) void
setBias
(int layer, int output, double bias) private void
(package private) void
setWeight
(int layer, int input, int output, double weight) (package private) Structure2D[]
-
Field Details
-
myBatchSize
private final int myBatchSize -
myInput
-
myNetwork
-
myOutputs
-
-
Constructor Details
-
WrappedANN
WrappedANN(ArtificialNeuralNetwork network, int batchSize)
-
-
Method Details
-
equals
-
get
- Specified by:
get
in interfaceSupplier<ArtificialNeuralNetwork>
-
hashCode
public int hashCode() -
newInputBatch
When usingNetworkTrainer
orNetworkInvoker
with a batch size larger than 1 this utility may help with creating the batches. -
setInput
-
adjust
void adjust(int layer, PhysicalStore<Double> input, PhysicalStore<Double> output, PhysicalStore<Double> upstreamGradient, PhysicalStore<Double> downstreamGradient) -
depth
int depth() -
getActivator
-
getBatchSize
int getBatchSize() -
getBias
double getBias(int layer, int output) -
getInput
PhysicalStore<Double> getInput() -
getInput
-
getOutput
PhysicalStore<Double> getOutput() -
getOutput
-
getOutputActivator
ArtificialNeuralNetwork.Activator getOutputActivator() -
getWeight
double getWeight(int layer, int input, int output) -
getWeights
List<MatrixStore<Double>> getWeights() -
invoke
-
newOutputBatch
DataBatch newOutputBatch() -
randomise
void randomise() -
setActivator
-
setBias
void setBias(int layer, int output, double bias) -
setWeight
void setWeight(int layer, int input, int output, double weight) -
structure
Structure2D[] structure()
-