Core ML Concepts Playground
Interactive visualizations of fundamental machine learning concepts
Project Overview
This project provides interactive visualizations and demonstrations of core machine learning concepts. Through hands-on exploration, users can gain intuitive understanding of fundamental ML principles including algorithms, gradient descent, neural networks, optimization, and more.
Everything coded from scratch!
Key Features
- Regression with real-time line fitting
- Classification algorithms comparison (k-NN, Decision Trees, SVM, Perceptron, etc.)
- PCA and dimensionality reduction visualization
- Clustering algorithms explorer (K-means, DBSCAN, Hierarchical, etc.)
- Gradient descent optimization visualization
- Bias-variance tradeoff demonstration
- Feature selection and importance analysis
- Neural network playground for building simple models
- Model evaluation and metrics
Interactive Demonstrations
Explore core ML concepts through hands-on visualizations. Click on any concept below to start learning!
Regression
Fit lines to data using a variety of regression methods, loss functoins, regularization, and optimization algorithms
Classification Algorithms
Explore k-NN, Decision Trees, SVM, and Perceptron
Dimensionality Reduction
Visualize PCA and understand data compression
Clustering Visualizer
Explore K-means, DBSCAN, and hierarchical clustering
Gradient Descent
Watch optimization algorithms navigate loss landscapes
Bias-Variance Tradeoff
Understand the fundamental tradeoff in model selection
Feature Selection
Discover which features matter most in your models
Neural Network Playground
Build and train simple neural networks interactively
Model Evaluation & Metrics
Explore accuracy, precision, recall, ROC curves, and confusion matrices