SmartEngine.NeuralNetworks.NeuronLayer Class Reference

A neuron layer is the trainable unit in the neural network graph. These can chained together to allow the network to learn and represent complex relationships. More...

Inheritance diagram for SmartEngine.NeuralNetworks.NeuronLayer:

Classes
class	CInfo

Public Member Functions
	NeuronLayer (CInfo cinfo)
NeuronLayer	InstanceLayer (string name, GraphNode input)
	Creates a new NeuronLayer with a different input, but weights shared with this layer. More...
void	SetRandomWeights ()
	Initialize the weights of the layer to random values. More...
void	Reset ()
	Resets the internal state. Should be called periodically on layers that need stepping (LSTM). The period to call this is defined by the sequence length used to train the layer. More...
void	Step ()
	Steps the internal state. Should be after execution of the layer and before new data is put into the layer. Only required to be called for layers that need stepping (LSTM). More...
unsafe SerializationResult	Serialize (MemoryBuffer buffer)
	Saves the data store to a buffer. More...
SerializationResult	Deserialize (MemoryBuffer buffer)
	Loads the data store and population from a buffer. More...
Public Member Functions inherited from SmartEngine.NeuralNetworks.GraphNode
Matrix	Execute ()
	Synchronously execute the graph. The returned matrix is read-only. More...
IEnumerator	RetrieveOutputAsync ()
	Asynchronously execute the graph. The output of the graph can be retrieved with the LastExecutionResult property when the returned enumerator completes. More...
Public Member Functions inherited from SmartEngine.Object
void	AddRef ()
	Increments the internal reference count on this object. It is not common to use this method directly. More...
void	Release ()
	Decrements the internal reference count on this object. It is not common to use this method directly. More...
Public Member Functions inherited from SmartEngine.Disposable
void	Dispose ()
	Cleans up any internal state. It is not safe to use an object after it has been disposed. More...

Additional Inherited Members
Properties inherited from SmartEngine.NeuralNetworks.GraphNode
string	Name [get]
	Returns the name of the graph node. More...
Matrix	LastExecutionResult [get]
	Get the results of last execution of the graph. The returned matrix is read-only. More...
int	OutputDimension [get]
	Returns the dimension of a single row of this node's output. The Output property's length will be a multiple of this value. More...

Detailed Description

A neuron layer is the trainable unit in the neural network graph. These can chained together to allow the network to learn and represent complex relationships.

Member Function Documentation

Deserialize()

SerializationResult SmartEngine.NeuralNetworks.NeuronLayer.Deserialize

(

MemoryBuffer

buffer

)

Loads the data store and population from a buffer.

Parameters

data

Returns

Implements SmartEngine.ISerializable.

InstanceLayer()

NeuronLayer SmartEngine.NeuralNetworks.NeuronLayer.InstanceLayer	(	string	name,
		GraphNode	input
	)

Creates a new NeuronLayer with a different input, but weights shared with this layer.

Parameters

name	The name for the new layer.
input	The input into the new layer.

Returns

Reset()

void SmartEngine.NeuralNetworks.NeuronLayer.Reset

(

)

Resets the internal state. Should be called periodically on layers that need stepping (LSTM). The period to call this is defined by the sequence length used to train the layer.

Serialize()

unsafe SerializationResult SmartEngine.NeuralNetworks.NeuronLayer.Serialize

(

MemoryBuffer

buffer

)

Saves the data store to a buffer.

Parameters

outData

Returns

Implements SmartEngine.ISerializable.

SetRandomWeights()

void SmartEngine.NeuralNetworks.NeuronLayer.SetRandomWeights

(

)

Initialize the weights of the layer to random values.

Step()

void SmartEngine.NeuralNetworks.NeuronLayer.Step

(

)

Steps the internal state. Should be after execution of the layer and before new data is put into the layer. Only required to be called for layers that need stepping (LSTM).