SpiralNetOptions
Options & Configuration · AiDotNet.Models.Options
Configuration options for SpiralNet++ mesh neural network.
Inheritance: ModelOptions → NeuralNetworkOptions → SpiralNetOptions
For Beginners
These options control how the SpiralNet++ network processes 3D mesh data. Key settings include:
- SpiralLength: How many neighbors to consider for each vertex
- ConvChannels: Feature sizes at each layer
- PoolRatios: How much to simplify the mesh at each pooling step
How It Works
SpiralNet++ is a mesh convolution architecture that uses spiral sequences to define consistent neighbor orderings on irregular mesh vertices.
Constructors
| Constructor | Summary |
|---|---|
SpiralNetOptions() |
Properties
| Property | Summary |
|---|---|
bool IncludeCoordinates { get; set; } |
Gets or sets whether to include vertex coordinates as input features. |
bool IncludeNormals { get; set; } |
Gets or sets whether to include vertex normals as input features. |
bool UseBatchNorm { get; set; } |
Gets or sets whether to use batch normalization. |
bool UseGlobalAveragePooling { get; set; } |
Gets or sets whether to use global average pooling before classification. |
double DropoutRate { get; set; } |
Gets or sets the dropout rate for regularization. |
double[] PoolRatios { get; set; } |
Gets or sets the pooling ratios for mesh simplification. |
int InputFeatures { get; set; } |
Gets or sets the number of input features per vertex. |
int NumClasses { get; set; } |
Gets or sets the number of output classes for classification. |
int NumVertices { get; set; } |
Gets or sets the default vertex count used to synthesize fallback spiral indices when explicit mesh topology is not provided. |
int SpiralLength { get; set; } |
Gets or sets the length of the spiral sequence for convolutions. |
int[] ConvChannels { get; set; } |
Gets or sets the channel sizes for each convolution layer. |
int[] FullyConnectedSizes { get; set; } |
Gets or sets the sizes of fully connected layers before output. |