ConstellationNetOptions<T, TInput, TOutput>
Options & Configuration · AiDotNet.MetaLearning.Options
Configuration options for ConstellationNet (Xu et al., ICLR 2021).
Inheritance: ModelOptions → ConstellationNetOptions<T, TInput, TOutput>
Implements: IMetaLearnerOptions<T>
For Beginners
ConstellationNet finds patterns in how parts relate:
- Detect discriminative parts/regions in each example
- Learn how these parts are arranged (their “constellation”)
- Compare queries to support classes by matching constellation patterns
Imagine recognizing a face: you don’t just look at eyes, nose, mouth separately. You also look at HOW they’re arranged - that’s the constellation.
How It Works
ConstellationNet learns structured representations where parts and their spatial relationships form “constellations” for few-shot classification.
Reference: Xu, C., Fu, Y., Liu, C., Wang, C., Li, J., Huang, F., Zhang, L., & Xue, X. (2021). Learning Dynamic Alignment via Meta-filter for Few-shot Learning. CVPR 2021.
Constructors
| Constructor | Summary |
|---|---|
ConstellationNetOptions(IFullModel<T, TInput, TOutput> metaModel) |
Initializes a new instance of ConstellationNetOptions. |
Properties
| Property | Summary |
|---|---|
IEpisodicDataLoader<T, TInput, TOutput> DataLoader { get; set; } |
Gets or sets the episodic data loader. |
IFullModel<T, TInput, TOutput> MetaModel { get; set; } |
Gets or sets the feature extractor model. |
IGradientBasedOptimizer<T, TInput, TOutput> InnerOptimizer { get; set; } |
Must stay null: this algorithm takes no plain gradient inner step (it adapts without one, or by its own update rule), so a supplied optimizer is rejected when the learner is built. |
IGradientBasedOptimizer<T, TInput, TOutput> MetaOptimizer { get; set; } |
Optimizer for the meta-update. |
ILossFunction<T> LossFunction { get; set; } |
Loss function for meta-training; null uses the default multi-class classification loss. |
bool EnableCheckpointing { get; set; } |
|
bool UseFirstOrder { get; set; } |
|
bool UseSpatialRelations { get; set; } |
Gets or sets whether to use spatial relationships between parts. |
double InnerLearningRate { get; set; } |
|
double OuterLearningRate { get; set; } |
|
double? GradientClipThreshold { get; set; } |
|
int AdaptationSteps { get; set; } |
Number of gradient steps taken on each task’s support set. |
int CheckpointFrequency { get; set; } |
|
int EvaluationFrequency { get; set; } |
|
int EvaluationTasks { get; set; } |
|
int MetaBatchSize { get; set; } |
|
int NumMetaIterations { get; set; } |
|
int NumParts { get; set; } |
Gets or sets the number of constellation parts to detect. |
int PartFeatureDim { get; set; } |
Gets or sets the part feature dimension. |
int? RandomSeed { get; set; } |
Gets or sets the random seed. |
Methods
| Method | Summary |
|---|---|
IMetaLearnerOptions<T> Clone() |
|
bool IsValid() |