ML Visualizations Hub
Interactive machine learning algorithm visualizations and tutorials
# Interactive learning through visual algorithms
algorithms = ["decision_trees", "clustering", "regression"]
for algo in algorithms:
visualize(algo, interactive=True)
Decision Tree Visualizer
Build, visualize, and make predictions with interactive decision trees. Explore supervised learning through multiple datasets and see how decisions are made at each node.
- Multiple datasets (Iris, Wine, etc.)
- Interactive tree visualization
- Real-time predictions
- Path highlighting
- Confidence scores
K-means Clustering
Discover patterns in data through unsupervised learning. Watch centroids move and see how K-means groups similar data points together step by step.
- Study & Interactive modes
- Elbow method for optimal K
- Animated clustering process
- Performance labeling
- Student data analysis
DBSCAN Clustering
Density-based clustering that discovers clusters of varying shapes and automatically detects outliers. No need to specify number of clusters beforehand.
- Interactive point placement
- Step-by-step visualization
- Automatic noise detection
- Parameter adjustment
- Core/Border/Noise classification
Linear Regression
Understand the fundamentals of regression analysis. Visualize how linear models fit data and predict continuous values with interactive examples.
- Interactive data point addition
- Least squares & gradient descent
- Real-time model metrics
- Prediction functionality
- Multiple sample datasets
Neural Networks
Dive deep into artificial neural networks. See how neurons process information and learn complex patterns through backpropagation.
- Interactive network builder
- Forward & backward propagation
- Multiple activation functions
- Real-time training animation
- 5 challenging datasets
Support Vector Machines
Explore the power of SVMs in finding optimal decision boundaries. Understand kernels and how SVMs handle non-linear data.
- Interactive point placement
- Multiple kernel functions
- Support vector highlighting
- Decision boundary visualization
- Kernel comparison tool
Reinforcement Learning Playground
Experiment with interactive reinforcement learning environments. Watch agents learn optimal strategies through trial and error, rewards, and exploration.
- Live agent training
- Multiple RL algorithms
- Customizable environments
- Reward visualization
- Step-by-step exploration
Random Forest
See how ensemble methods combine multiple decision trees to create more robust and accurate predictions through democratic voting.
- Forest visualization
- Bootstrap sampling
- Feature randomness
- Voting mechanisms
- Feature importance
1. explore_decision_trees() # Understand supervised learning
2. master_regression() # Predict continuous values
3. explore_svm() # Maximum margin classification
4. build_neural_nets() # Deep learning fundamentals
5. discover_kmeans() # Learn unsupervised clustering
6. master_dbscan() # Density-based clustering with noise
7. experiment_rl() # Interactive reinforcement learning
8. soon_random_forest() # Each visualization builds upon the previous concepts