GANDALFClassifier<T>

Models & Types ยท AiDotNet.NeuralNetworks.Tabular

GANDALF implementation for classification tasks.

Inheritance: GANDALFBase<T> โ†’ GANDALFClassifier<T>

For Beginners

Use GANDALF for classification when you want:

  • Automatic feature importance learning
  • Interpretable predictions (can see feature weights and tree paths)
  • Good performance on tabular data

Example:

How It Works

GANDALFClassifier uses gated feature selection with neural decision trees for multi-class classification. The additive ensemble of trees produces class logits that are converted to probabilities via softmax.

Constructors

Constructor Summary
GANDALFClassifier(int numFeatures, int numClasses, GANDALFOptions<T> options) Initializes a new instance of the GANDALFClassifier class.

Properties

Property Summary
int NumClasses { get; } Gets the number of output classes.

Methods

Method Summary
T ComputeAccuracy(Tensor<T> features, Vector<int> targets) Computes classification accuracy.
T ComputeCrossEntropyLoss(Tensor<T> probabilities, Vector<int> targets) Computes the cross-entropy loss.
T TrainStep(Tensor<T> features, Vector<int> targets, T learningRate) Performs a single training step.
Tensor<T> Forward(Tensor<T> features) Performs the forward pass to get class logits.
Tensor<T> PredictProbabilities(Tensor<T> features) Predicts class probabilities using softmax.
Vector<int> Predict(Tensor<T> features) Predicts the most likely class for each sample.
void ResetState() Resets internal state.
void UpdateParameters(T learningRate) Updates all parameters.