Classical ML

Complete reference for all 106+ classical ML algorithms in AiDotNet.

Classification (28 Algorithms)

Linear Models

Algorithm	Description	Multi-class
`LogisticRegression<T>`	Binary/multi-class logistic	Yes
`LinearSVC<T>`	Linear Support Vector Classifier	Yes
`RidgeClassifier<T>`	Ridge regression for classification	Yes
`SGDClassifier<T>`	Stochastic Gradient Descent	Yes
`Perceptron<T>`	Linear perceptron	Yes

var classifier = new LogisticRegression<double>(
    regularization: 0.01,
    solver: LogisticSolver.LBFGS,
    maxIterations: 1000);

Support Vector Machines

Algorithm	Description	Kernel
`SVC<T>`	Support Vector Classifier	Multiple
`NuSVC<T>`	Nu-Support Vector Classifier	Multiple
`LinearSVC<T>`	Linear SVC	Linear

Supported kernels: Linear, Polynomial, RBF, Sigmoid, Precomputed

var svm = new SVC<double>(
    kernel: KernelType.RBF,
    C: 1.0,
    gamma: "scale");

Tree-Based

Algorithm	Description
`DecisionTreeClassifier<T>`	Single decision tree
`RandomForestClassifier<T>`	Ensemble of trees
`ExtraTreesClassifier<T>`	Extremely randomized trees
`GradientBoostingClassifier<T>`	Gradient boosting
`XGBoostClassifier<T>`	XGBoost implementation
`LightGBMClassifier<T>`	LightGBM implementation
`CatBoostClassifier<T>`	CatBoost implementation

var forest = new RandomForestClassifier<double>(
    nEstimators: 100,
    maxDepth: 10,
    minSamplesSplit: 2);

Naive Bayes

Algorithm	Distribution
`GaussianNB<T>`	Gaussian (continuous)
`MultinomialNB<T>`	Multinomial (counts)
`BernoulliNB<T>`	Bernoulli (binary)
`ComplementNB<T>`	Complement (imbalanced)

Neighbors

Algorithm	Description
`KNeighborsClassifier<T>`	K-Nearest Neighbors
`RadiusNeighborsClassifier<T>`	Radius-based neighbors

Neural Networks

Algorithm	Description
`MLPClassifier<T>`	Multi-layer Perceptron

Ensemble

Algorithm	Description
`BaggingClassifier<T>`	Bootstrap aggregating
`AdaBoostClassifier<T>`	Adaptive Boosting
`VotingClassifier<T>`	Soft/hard voting
`StackingClassifier<T>`	Stacked generalization

Regression (41 Algorithms)

Linear Models

Algorithm	Description
`LinearRegression<T>`	Ordinary Least Squares
`Ridge<T>`	L2 regularization
`Lasso<T>`	L1 regularization
`ElasticNet<T>`	L1 + L2 regularization
`Lars<T>`	Least Angle Regression
`LassoLars<T>`	LARS with L1
`OrthogonalMatchingPursuit<T>`	Sparse approximation
`BayesianRidge<T>`	Bayesian regression
`ARDRegression<T>`	Automatic Relevance Determination
`SGDRegressor<T>`	SGD for regression
`HuberRegressor<T>`	Robust to outliers
`RANSACRegressor<T>`	RANSAC robust regression
`TheilSenRegressor<T>`	Theil-Sen estimator
`PassiveAggressiveRegressor<T>`	Online learning

var model = new ElasticNet<double>(
    alpha: 1.0,
    l1Ratio: 0.5,
    maxIterations: 1000);

Support Vector Regression

Algorithm	Description
`SVR<T>`	Support Vector Regression
`NuSVR<T>`	Nu-Support Vector Regression
`LinearSVR<T>`	Linear SVR

Tree-Based

Algorithm	Description
`DecisionTreeRegressor<T>`	Single decision tree
`RandomForestRegressor<T>`	Ensemble of trees
`ExtraTreesRegressor<T>`	Extremely randomized trees
`GradientBoostingRegressor<T>`	Gradient boosting
`XGBoostRegressor<T>`	XGBoost implementation
`LightGBMRegressor<T>`	LightGBM implementation
`CatBoostRegressor<T>`	CatBoost implementation

Neighbors

Algorithm	Description
`KNeighborsRegressor<T>`	K-Nearest Neighbors
`RadiusNeighborsRegressor<T>`	Radius-based neighbors

Gaussian Processes

Algorithm	Description
`GaussianProcessRegressor<T>`	GP regression

Neural Networks

Algorithm	Description
`MLPRegressor<T>`	Multi-layer Perceptron

Ensemble

Algorithm	Description
`BaggingRegressor<T>`	Bootstrap aggregating
`AdaBoostRegressor<T>`	Adaptive Boosting
`VotingRegressor<T>`	Average predictions
`StackingRegressor<T>`	Stacked generalization

Isotonic

Algorithm	Description
`IsotonicRegression<T>`	Monotonic regression

Quantile

Algorithm	Description
`QuantileRegressor<T>`	Quantile regression

Clustering (20+ Algorithms)

Centroid-Based

Algorithm	Description
`KMeans<T>`	K-Means clustering
`MiniBatchKMeans<T>`	Mini-batch K-Means
`KMedoids<T>`	K-Medoids (PAM)
`BisectingKMeans<T>`	Bisecting K-Means

var kmeans = new KMeans<double>(
    nClusters: 5,
    maxIterations: 300,
    init: KMeansInit.KMeansPlusPlus);

Density-Based

Algorithm	Description
`DBSCAN<T>`	Density-based spatial clustering
`HDBSCAN<T>`	Hierarchical DBSCAN
`OPTICS<T>`	Ordering Points for clustering
`MeanShift<T>`	Mean shift clustering

Hierarchical

Algorithm	Description
`AgglomerativeClustering<T>`	Bottom-up hierarchical
`BIRCH<T>`	Balanced Iterative Reducing
`Ward<T>`	Ward’s minimum variance

Spectral

Algorithm	Description
`SpectralClustering<T>`	Graph-based clustering
`SpectralBiclustering<T>`	Biclustering
`SpectralCoclustering<T>`	Co-clustering

Model-Based

Algorithm	Description
`GaussianMixture<T>`	GMM clustering
`BayesianGaussianMixture<T>`	Bayesian GMM

Other

Algorithm	Description
`AffinityPropagation<T>`	Message passing
`CLARA<T>`	Clustering Large Applications
`CLARANS<T>`	Randomized CLARA

Dimensionality Reduction

Linear Methods

Algorithm	Description
`PCA<T>`	Principal Component Analysis
`IncrementalPCA<T>`	Online PCA
`KernelPCA<T>`	Kernel PCA
`SparsePCA<T>`	Sparse PCA
`TruncatedSVD<T>`	Truncated SVD
`FactorAnalysis<T>`	Factor Analysis
`FastICA<T>`	Independent Component Analysis
`NMF<T>`	Non-negative Matrix Factorization
`LatentDirichletAllocation<T>`	Topic modeling

Manifold Learning

Algorithm	Description
`TSNE<T>`	t-SNE visualization
`UMAP<T>`	Uniform Manifold Approximation
`Isomap<T>`	Isometric Mapping
`LocallyLinearEmbedding<T>`	LLE
`MDS<T>`	Multidimensional Scaling
`SpectralEmbedding<T>`	Spectral embedding

Anomaly Detection

Algorithm	Description
`IsolationForest<T>`	Isolation-based
`LocalOutlierFactor<T>`	Density-based
`OneClassSVM<T>`	One-class SVM
`EllipticEnvelope<T>`	Gaussian distribution
`ECOD<T>`	Empirical Cumulative Distribution

Usage with AiModelBuilder

// Classification
var result = await new AiModelBuilder<double, double[], int>()
    .ConfigureModel(new RandomForestClassifier<double>(nEstimators: 100))
    .ConfigurePreprocessing()
    .BuildAsync(features, labels);

// Regression
var result = await new AiModelBuilder<double, double[], double>()
    .ConfigureModel(new GradientBoostingRegressor<double>())
    .ConfigurePreprocessing()
    .BuildAsync(features, targets);

// Clustering
var result = await new AiModelBuilder<double, double[], int>()
    .ConfigureModel(new HDBSCAN<double>(minClusterSize: 5))
    .BuildAsync(features);