Learn machine learning
Ten stops, in order: the eight topic planets from the home page plus two overviews. Start with data preparation on the left, then follow the lines: supervised learning above, unsupervised below, and the systems built on them to the right. Every page has a playground.
-
01 Data preparation
FoundationsMost of the work, and most of the mistakes, happen before any model is trained. How data is prepared for each kind of learning.
-
02 Supervised learning
SupervisedLearning a mapping from inputs to a known answer. Regression predicts numbers, classification predicts categories.
-
03 Regression
SupervisedPredicting a number. From a straight line to boosted trees, and how to tell a good fit from a memorised one.
-
04 Classification
SupervisedPredicting a category, and a probability for it. Decision boundaries, thresholds and metrics that survive imbalance.
-
05 Unsupervised learning
UnsupervisedFinding structure when nobody has labelled the data. Clustering groups rows, dimensionality reduction compresses columns.
-
06 Clustering
UnsupervisedGrouping points that belong together, without labels. Centroids, densities, mixtures and hierarchies.
-
07 Dimensionality reduction
UnsupervisedSqueezing many features into a few while keeping what matters, for plotting, compression and cleaner inputs.
-
08 Embeddings
RepresentationTurning text into vectors so that similar meaning becomes small distance. The foundation of semantic search and RAG.
-
09 Retrieval-augmented generation
SystemsGiving a language model the right documents at the right moment: the full pipeline from ingestion to evaluation.
-
10 Recommender systems
SystemsPredicting what each person will want next, from what they and people like them did before.