Linear Regression Visualizer
Fit a line, inspect residuals, and see gradient descent reduce squared error.
Interactive machine learning, deep learning & AI visualizations.
Choose one focused guide to inspect real calculations, model states, visual evidence, formulas, and the effect of changing parameters.
Start Guided Learning →Learn the intuition first, then verify it in the real lab.
Guided Learning uses short lessons, selectable example styles, quick checks, and direct links into the existing ModelOrigen calculations.
Machine Learning & Deep Learning
Logistic Regression Visualizer
Turn a weighted score into probability and watch the decision boundary move.
K-Means Visualizer
Assign points, move centroids, and inspect inertia through real iterations.
K-Nearest Neighbors Visualizer
See distances, the nearest K samples, votes, and the local decision boundary.
Decision Tree Visualizer
Watch candidate thresholds compete, then follow a sample from root to leaf.
Support Vector Machine Visualizer
Inspect decision scores, margins, hinge loss, kernels, and active support influence.
Random Forest Visualizer
Train diverse trees, inspect their votes, and see the aggregate decision stabilize.
Gradient Boosting Visualizer
Add weak trees sequentially and watch each stage correct the current ensemble.
XGBoost Visualizer
See gradients, Hessians, regularized split gain, leaf scores, and eta-scaled trees.
AdaBoost Visualizer
Watch difficult samples gain influence and weak learners receive weighted votes.
Extra Trees Visualizer
Increase tree randomness, then inspect how many randomized trees aggregate their predictions.
Neural Network / MLP Visualizer
Build a neural network playground in the browser: add/remove hidden layers and neurons, train real backpropagation, and replay how weights and the decision boundary change.
Convolutional Neural Network Visualizer
Explore a tiny genuinely trained CNN with two learned 3×3 filters, ReLU feature maps, max pooling, an eight-feature dense classifier, and real kernel gradients.
Recurrent Neural Network Visualizer
Train a tiny RNN with genuine hidden states and backpropagation through time.
Long Short-Term Memory Visualizer
Train and inspect a tiny LSTM with real forget, input, candidate, output, cell, and hidden-state calculations.
Gated Recurrent Unit Visualizer
Train and inspect a tiny GRU with genuine update gate, reset gate, candidate, and recurrent gradients.
Transformer Self-Attention Visualizer
Inspect genuine Q, K, V projections, scaled dot-product scores, stable softmax weights, and attention outputs.
Autoencoder Visualizer
Train a tiny browser-side autoencoder and inspect the bottleneck, reconstructions, loss, and real gradients.
Generative Adversarial Network Visualizer
Train a tiny 1D browser-side GAN and inspect genuine generator/discriminator losses and distribution changes.
Data Science & AI Foundations
Statistics & Sampling
Watch one real sample become one sample mean, then one point in a growing sampling distribution.
Principal Component Analysis
Reveal PCA in calculation order: raw data → centering/scaling → covariance → eigenvectors → projection.
Gradient Sampling Lab
Compare full-batch, mini-batch, and one-example stochastic gradients on one genuine loss surface.
Optimizer Studio
Watch SGD, Momentum, RMSProp, and Adam descend the same 3D loss surface with the same mini-batch evidence.
Bias–Variance Studio
Move model complexity and watch fit, train error, validation error, bias², and variance change together.
Explainable AI Lab
Freeze one fitted model, then inspect permutation importance, PDP, ICE, and exact local model terms while keeping the model and its explanations clearly separate.
Time Series Lab
Keep chronology visible while trend, seasonality, lag relationships, smoothing, train/test cutoffs, and one-step forecasts change.
Model Evaluation Studio
Move the decision threshold and watch individual held-out scores become TP, FP, TN, FN, metrics, ROC/PR operating points, and calibration evidence.
Cross-Validation Visualizer
Watch k-fold validation rotate: split → refit on k−1 folds → predict the held-out fold → score → repeat → aggregate.
Data Cleaning Lab
Trace raw rows through diagnosis, imputation, duplicate handling, IQR outlier evidence, scaling, and categorical encoding.
Feature Engineering Studio
Keep the learner fixed and change only φ(x): raw x, standardized x, x², or bucket indicators—then inspect the design row and prediction terms.
Probability & Bayes Lab
Watch prior belief branch into competing likelihoods, keep only paths compatible with evidence, normalize them, and arrive at the posterior.