Machine Learning, Not Just Theory
Change parameters, run a model, and watch data become decisions in real time. No setup, no code in the way.
Tutorials, experiments, and AI updates delivered to your inbox every week.
Join hundreds of learners exploring machine learning visually.
Written and curated by experienced machine learning researchers.
Start with a strong read.
Read the idea first, then open the related lab when you are ready to experiment.
AI Basics20 Important AI Concepts We Can Understand in 20 Minutes
A clean, beginner-friendly guide to the core ideas behind modern AI, from neural networks and transformers to RAG, agents, and diffusion models.
Updated Jul 31, 2026 ยท 12 min read
Dimensionality ReductionPCA Explained: How Machine Learning Simplifies Data Without Losing the Main Story11 min read
OptimizationGradient Descent Explained: A Visual and Intuitive Guide12 min read
EvaluationConfusion Matrix and Classification Metrics10 min read
Deep LearningTransformer and Attention Mechanism: From the Paper to an Interactive Lab14 min readReal models for real use cases.
Project · YOLOBlood Cell CounterDetect and count blood cells from microscope images.
Application · Computer VisionTry the Blood Cell CounterUpload an image and run the detection model directly.
Application · ClassificationObesity Risk DetectionRun a classification model on lifestyle and body data.
Application · Remote SensingForest NDVI MappingCompare red and NIR bands to read vegetation health.
Project · Dimensionality ReductionPCA and LDA ClassificationExplore feature projections before the model separates classes.
Pick a concept and try it live.
Big visuals, manual data input, and step-by-step animation to understand every algorithm.
PCA
Choose 2 or 3 principal components and compare 2D and 3D PCA projections based on explained variance.
Z = XW
OptimizationOpen
Gradient Descent
Pick a starting point, compare learning rates, and follow the path toward a local or global minimum.
theta <- theta - alpha * grad L
EnsembleOpen
Random Forest
Compare votes across many trees and see the ensemble decision boundary become more stable.
y = mode(h_1(x), ..., h_T(x))
Deep LearningOpen
Neural Network
Configure hidden layers, train an XOR network, and read every neuron activation in the architecture.
a = sigma(Wx + b)
Deep LearningOpen
CNN Convolution
Edit input pixels, choose a kernel, and see the feature map computed through convolution.
Y[i,j] = sum_m sum_n X[i+m,j+n] K[m,n] + b
Deep LearningOpen
Transformer
Type your own sentence and read Q, K, V, scaled dot-product, softmax, and the attention matrix.
Attention(Q,K,V) = softmax(QK^T / sqrt(d))V
Frequency DomainOpen
FFT Spectrum
Enter your own signal or pick an example, then map the waveform to an FFT spectrum up to Nyquist.
X[k] = sum x[n] exp(-j 2 pi kn / N)