HumachLearnLearn · explore · apply
Start learning
Interactive model reference

Understand the mechanism—not just the API.

Explore model intuition, conceptual training animations, synchronized Reset/Train/Pause/Train Step controls, 2D/3D behaviour, detailed explanations, evaluation guidance and copyable Python examples with output.

Supervised Learning19 models
Binary Classification2 model page(s)2
Logistic RegressionLogistic Regression is a probabilistic linear classifier that estimates the probability of a binary outcome. It is widely used as an interpretable baseline and as a production model wheClassificationSupport Vector Classifier (SVC)Support Vector Classification separates classes by maximizing the margin between them. Kernel functions can extend the boundary beyond a linear hyperplane.Classification
Binary / Multi-Class1 model page(s)1
Gaussian Naive BayesGaussian Naive Bayes is a supervised learning method in the binary / multi-class family. This page summarizes its mechanism, practical uses, important trade-offs, and a browser-based coClassification
Multi-Class Classification3 model page(s)3
Decision Tree Classifier (CART)Decision Tree Classifier (CART) applies the Decision Trees (CART) learning mechanism to categorical targets. Decision Trees (CART) is a supervised learning method in the multi-class claClassificationSoftmax Regression (Multinomial Logistic)Softmax Regression (Multinomial Logistic) is a supervised learning method in the multi-class classification family. This page summarizes its mechanism, practical uses, important trade-oClassificationK-Nearest Neighbors (KNN) ClassifierK-Nearest Neighbors (KNN) Classifier applies the K-Nearest Neighbors (KNN) learning mechanism to categorical targets. K-Nearest Neighbors (KNN) is a supervised learning method in the muClassification
Non-Linear Regression3 model page(s)3
Decision Tree Regressor (CART)Decision Tree Regressor (CART) applies the Decision Trees (CART) learning mechanism to continuous targets, producing numeric predictions instead of class labels.RegressionK-Nearest Neighbors (KNN) RegressorK-Nearest Neighbors (KNN) Regressor applies the K-Nearest Neighbors (KNN) learning mechanism to continuous targets, producing numeric predictions instead of class labels.RegressionSupport Vector Regressor (SVR)Support Vector Regression fits a function inside an ε-insensitive tube and penalizes observations that fall outside the tube, balancing flatness and prediction error.Regression
Multi-Label Classification2 model page(s)2
Binary RelevanceBinary Relevance is a supervised learning method in the multi-label classification family. This page summarizes its mechanism, practical uses, important trade-offs, and a browser-based ClassificationClassifier ChainsClassifier Chains is a supervised learning method in the multi-label classification family. This page summarizes its mechanism, practical uses, important trade-offs, and a browser-basedClassification
Linear Regression5 model page(s)5
Ridge Regression (L2)Ridge Regression (L2) is a supervised learning method in the linear regression family. This page summarizes its mechanism, practical uses, important trade-offs, and a browser-based concRegressionLasso Regression (L1)Lasso Regression (L1) is a supervised learning method in the linear regression family. This page summarizes its mechanism, practical uses, important trade-offs, and a browser-based concRegressionElasticNetElasticNet is a supervised learning method in the linear regression family. This page summarizes its mechanism, practical uses, important trade-offs, and a browser-based concept exploreRegressionOrdinary Least Squares (Linear Regression)Ordinary Least Squares (OLS) is the canonical linear regression model. It estimates coefficients that minimize squared prediction error and provides an interpretable baseline for continRegressionPolynomial RegressionExpands numeric inputs into polynomial and interaction features so a linear estimator can learn curved response surfaces.Regression
Time-Series Forecasting2 model page(s)2
ARIMA / SARIMAARIMA / SARIMA is a supervised learning method in the time-series forecasting family. This page summarizes its mechanism, practical uses, important trade-offs, and a browser-based conceForecastingProphetProphet is a supervised learning method in the time-series forecasting family. This page summarizes its mechanism, practical uses, important trade-offs, and a browser-based concept explForecasting
Probabilistic Classification1 model page(s)1
Gaussian Process Classifier (GPC)A Bayesian non-parametric classifier that places a Gaussian-process prior over a latent function and maps that latent function through a link function to class probabilities. It is espeClassification
Probabilistic Regression1 model page(s)1
Gaussian Process Regressor (GPR)A Bayesian non-parametric regression model that treats functions as random variables. Instead of returning only a fitted curve, GPR naturally provides a predictive mean and predictive uRegression
Unsupervised Learning9 models
Centroid-based Clustering1 model page(s)1
K-Means / K-Means++K-Means is a centroid-based unsupervised clustering algorithm that partitions observations into a user-specified number of compact groups. K-Means++ improves initialization so the startClustering
Density-based Clustering1 model page(s)1
DBSCANDBSCAN is an unsupervised learning method in the density-based clustering family. This page summarizes its mechanism, practical uses, important trade-offs, and a browser-based concept exClustering
Hierarchical Clustering1 model page(s)1
Agglomerative Hierarchical ClusteringAgglomerative Hierarchical Clustering is an unsupervised learning method in the hierarchical clustering family. This page summarizes its mechanism, practical uses, important trade-offs, Clustering
Distribution-based Clustering1 model page(s)1
Gaussian Mixture Models (GMM)Gaussian Mixture Models (GMM) is an unsupervised learning method in the distribution-based clustering family. This page summarizes its mechanism, practical uses, important trade-offs, anClustering
Linear Dimensionality Reduction1 model page(s)1
Principal Component Analysis (PCA)Principal Component Analysis (PCA) is an unsupervised learning method in the linear dimensionality reduction family. This page summarizes its mechanism, practical uses, important trade-oDimensionality Reduction
Non-Linear Manifold Learning2 model page(s)2
UMAPUMAP is an unsupervised learning method in the non-linear manifold learning family. This page summarizes its mechanism, practical uses, important trade-offs, and a browser-based concept Dimensionality Reductiont-SNEt-SNE is an unsupervised learning method in the non-linear manifold learning family. This page summarizes its mechanism, practical uses, important trade-offs, and a browser-based conceptDimensionality Reduction
Anomaly Detection2 model page(s)2
Isolation ForestIsolation Forest is an unsupervised learning method in the anomaly detection family. This page summarizes its mechanism, practical uses, important trade-offs, and a browser-based conceptAnomaly DetectionLocal Outlier Factor (LOF)Local Outlier Factor (LOF) is an unsupervised learning method in the anomaly detection family. This page summarizes its mechanism, practical uses, important trade-offs, and a browser-basAnomaly Detection
Semi-Supervised & Self-Supervised4 models
Semi-Supervised Learning2 model page(s)2
Pseudo-LabelingPseudo-Labeling is a semi-supervised & self-supervised method in the semi-supervised learning family. This page summarizes its mechanism, practical uses, important trade-offs, and a broRepresentation LearningFixMatchFixMatch is a semi-supervised & self-supervised method in the semi-supervised learning family. This page summarizes its mechanism, practical uses, important trade-offs, and a browser-baRepresentation Learning
Self-Supervised Learning2 model page(s)2
SimCLRSimCLR is a semi-supervised & self-supervised method in the self-supervised learning family. This page summarizes its mechanism, practical uses, important trade-offs, and a browser-baseRepresentation LearningMasked Autoencoders (MAE)Masked Autoencoders (MAE) is a semi-supervised & self-supervised method in the self-supervised learning family. This page summarizes its mechanism, practical uses, important trade-offs,Generative Modeling
Reinforcement Learning (RL)6 models
Value-Based RL1 model page(s)1
Deep Q-Networks (DQN)Deep Q-Networks (DQN) is a reinforcement learning (rl) method in the value-based rl family. This page summarizes its mechanism, practical uses, important trade-offs, and a browser-basedReinforcement Learning
Policy-Based RL1 model page(s)1
REINFORCEREINFORCE is a reinforcement learning (rl) method in the policy-based rl family. This page summarizes its mechanism, practical uses, important trade-offs, and a browser-based concept exReinforcement Learning
Actor-Critic RL2 model page(s)2
Proximal Policy Optimization (PPO)Proximal Policy Optimization (PPO) is a reinforcement learning (rl) method in the actor-critic rl family. This page summarizes its mechanism, practical uses, important trade-offs, and aReinforcement LearningSoft Actor-Critic (SAC)Soft Actor-Critic (SAC) is a reinforcement learning (rl) method in the actor-critic rl family. This page summarizes its mechanism, practical uses, important trade-offs, and a browser-baReinforcement Learning
Model-Based RL1 model page(s)1
MuZeroMuZero is a reinforcement learning (rl) method in the model-based rl family. This page summarizes its mechanism, practical uses, important trade-offs, and a browser-based concept explorReinforcement Learning
Alignment & Preference1 model page(s)1
Direct Preference Optimization (DPO)Direct Preference Optimization (DPO) is a reinforcement learning (rl) method in the alignment & preference family. This page summarizes its mechanism, practical uses, important trade-ofReinforcement Learning
Ensemble Learning & Modern Enablers13 models
Bagging1 model page(s)1
Random Forest ClassifierRandom Forest Classifier applies the Random Forest learning mechanism to categorical targets. Random Forest is a ensemble learning & modern enablers method in the bagging family. This pClassification
Non-Linear Regression5 model page(s)5
Random Forest RegressorRandom Forest Regressor applies the Random Forest learning mechanism to continuous targets, producing numeric predictions instead of class labels.RegressionXGBoost RegressorXGBoost Regressor applies the XGBoost learning mechanism to continuous targets, producing numeric predictions instead of class labels.RegressionLightGBM RegressorLightGBM Regressor applies the LightGBM learning mechanism to continuous targets, producing numeric predictions instead of class labels.RegressionCatBoost RegressorCatBoost Regressor applies the CatBoost learning mechanism to continuous targets, producing numeric predictions instead of class labels.RegressionStacking RegressorStacking Regressor applies the StackingClassifier / StackingRegressor learning mechanism to continuous targets, producing numeric predictions instead of class labels.Regression
Boosting3 model page(s)3
XGBoost ClassifierXGBoost Classifier applies the XGBoost learning mechanism to categorical targets. XGBoost is a ensemble learning & modern enablers method in the boosting family. This page summarizes itClassificationLightGBM ClassifierLightGBM Classifier applies the LightGBM learning mechanism to categorical targets. LightGBM is a ensemble learning & modern enablers method in the boosting family. This page summarizesClassificationCatBoost ClassifierCatBoost Classifier applies the CatBoost learning mechanism to categorical targets. CatBoost is a ensemble learning & modern enablers method in the boosting family. This page summarizesClassification
Stacking1 model page(s)1
Stacking ClassifierStacking Classifier applies the StackingClassifier / StackingRegressor learning mechanism to categorical targets. StackingClassifier / StackingRegressor is a ensemble learning & modern Classification
PEFT2 model page(s)2
LoRA (Low-Rank Adaptation)LoRA (Low-Rank Adaptation) is a ensemble learning & modern enablers method in the peft family. This page summarizes its mechanism, practical uses, important trade-offs, and a browser-baTraining SystemsQLoRAQLoRA is a ensemble learning & modern enablers method in the peft family. This page summarizes its mechanism, practical uses, important trade-offs, and a browser-based concept explorer.Training Systems
Distributed Systems1 model page(s)1
Fully Sharded Data Parallel (FSDP)Fully Sharded Data Parallel (FSDP) is a ensemble learning & modern enablers method in the distributed systems family. This page summarizes its mechanism, practical uses, important tradeTraining Systems
Deep Learning & Neural Architectures12 models
Core Neural Networks1 model page(s)1
Multilayer Perceptron (MLP) ClassifierMultilayer Perceptron (MLP) Classifier applies the Multilayer Perceptron (MLP) learning mechanism to categorical targets. Multilayer Perceptron (MLP) is a deep learning & neural architeClassification
Non-Linear Regression5 model page(s)5
Multilayer Perceptron (MLP) RegressorMultilayer Perceptron (MLP) Regressor applies the Multilayer Perceptron (MLP) learning mechanism to continuous targets, producing numeric predictions instead of class labels.RegressionResNet RegressorResNet Regressor applies the ResNet learning mechanism to continuous targets, producing numeric predictions instead of class labels.RegressionLSTM RegressorLSTM Regressor applies the Long Short-Term Memory (LSTM) learning mechanism to continuous targets, producing numeric predictions instead of class labels.RegressionMamba / S4 RegressorMamba / S4 Regressor applies the Mamba / S4 learning mechanism to continuous targets, producing numeric predictions instead of class labels.RegressionTransformer / Attention RegressorTransformer / Attention Regressor applies the Transformers & Attention Models (GPT / LLaMA / Claude / Gemini) learning mechanism to continuous targets, producing numeric predictions insRegression
Computer Vision2 model page(s)2
ResNet ClassifierResNet Classifier applies the ResNet learning mechanism to categorical targets. ResNet is a deep learning & neural architectures method in the computer vision family. This page summarizClassificationYOLO (v8 - v11)YOLO (v8 - v11) is a deep learning & neural architectures method in the computer vision family. This page summarizes its mechanism, practical uses, important trade-offs, and a browser-bObject Detection
Sequential Networks1 model page(s)1
LSTM ClassifierLSTM Classifier applies the Long Short-Term Memory (LSTM) learning mechanism to categorical targets. Long Short-Term Memory (LSTM) is a deep learning & neural architectures method in thClassification
State Space Models1 model page(s)1
Mamba / S4 ClassifierMamba / S4 Classifier applies the Mamba / S4 learning mechanism to categorical targets. Mamba / S4 is a deep learning & neural architectures method in the state space models family. ThiClassification
Transformers1 model page(s)1
Transformer / Attention ClassifierTransformer / Attention Classifier applies the Transformers & Attention Models (GPT / LLaMA / Claude / Gemini) learning mechanism to categorical targets. Transformers & Attention ModelsClassification
Generative Modeling1 model page(s)1
Stable Diffusion / Diffusion ModelsStable Diffusion / Diffusion Models is a deep learning & neural architectures method in the generative modeling family. This page summarizes its mechanism, practical uses, important traGenerative Modeling
Graph Machine Learning11 models
Graph Attention Networks2 model page(s)2
Graph Attention Network (GAT) ClassifierGraph Attention Network (GAT) Classifier applies the Graph Attention Networks (GAT) learning mechanism to categorical targets. Graph Attention Networks (GAT) is a deep learning & neuralClassificationGraph Attention Network (GAT) RegressorGraph Attention Network (GAT) Regressor applies the Graph Attention Networks (GAT) learning mechanism to continuous targets, producing numeric predictions instead of class labels.Regression
Graph Neural Networks9 model page(s)9
Graph Convolutional Network (GCN)Graph Convolutional Network (GCN) is a graph neural architecture for learning representations from entities connected by edges. Its interactive page visualises how information travels tGraph LearningGraphSAGEGraphSAGE is a graph neural architecture for learning representations from entities connected by edges. Its interactive page visualises how information travels through a graph, how a noGraph LearningGraph Isomorphism Network (GIN)Graph Isomorphism Network (GIN) is a graph neural architecture for learning representations from entities connected by edges. Its interactive page visualises how information travels thrGraph LearningGCNIIGCNII is a graph neural architecture for learning representations from entities connected by edges. Its interactive page visualises how information travels through a graph, how a node rGraph LearningGraph TransformerGraph Transformer is a graph neural architecture for learning representations from entities connected by edges. Its interactive page visualises how information travels through a graph, Graph LearningMessage Passing Neural Network (MPNN)Message Passing Neural Network (MPNN) is a graph neural architecture for learning representations from entities connected by edges. Its interactive page visualises how information traveGraph LearningChebyshev Graph Convolution (ChebNet)Chebyshev Graph Convolution (ChebNet) is a graph neural architecture for learning representations from entities connected by edges. Its interactive page visualises how information traveGraph LearningAPPNPAPPNP is a graph neural architecture for learning representations from entities connected by edges. Its interactive page visualises how information travels through a graph, how a node rGraph LearningSimplified Graph Convolution (SGC)Simplified Graph Convolution (SGC) is a graph neural architecture for learning representations from entities connected by edges. Its interactive page visualises how information travels Graph Learning