Class DiffWaveNetwork<T>

Namespace

AiDotNet.Diffusion.Models

Assembly

AiDotNet.dll

DiffWave neural network with dilated convolutions.

public class DiffWaveNetwork<T>

Type Parameters

T

The numeric type.

Inheritance

object

DiffWaveNetwork<T>

Inherited Members

object.Equals(object)

object.Equals(object, object)

object.GetHashCode()

object.GetType()

object.MemberwiseClone()

object.ReferenceEquals(object, object)

object.ToString()

Constructors

DiffWaveNetwork(int, int, int, int, int?)

Initializes a new DiffWaveNetwork.

public DiffWaveNetwork(int residualChannels = 64, int residualLayers = 30, int dilationCycle = 10, int melChannels = 80, int? seed = null)

Parameters

residualChannels int

residualLayers int

dilationCycle int

melChannels int

seed int?

Properties

ParameterCount

Gets the number of parameters.

public int ParameterCount { get; }

Property Value

int

Methods

Forward(Tensor<T>, int, Tensor<T>?)

Forward pass through the network.

public Tensor<T> Forward(Tensor<T> audio, int timestep, Tensor<T>? melCondition)

Parameters

audio Tensor<T>

timestep int

melCondition Tensor<T>

Returns

Tensor<T>

GetParameters()

Gets all parameters.

public Vector<T> GetParameters()

Returns

Vector<T>

SetParameters(Vector<T>)

Sets all parameters.

public void SetParameters(Vector<T> parameters)

Parameters

parameters Vector<T>