Ml

Learn Machine Learning: Doing It Honestly

The judgement that decides whether a model works: baselines, errors that count, seeds and margins, fair comparisons, the many ways test data leaks in, wrong labels and fairness, and reporting a result others can check.

Start Machine Learning: Doing It Honestly free → Browse all subjects

Free to start · adaptive placement finds your level · reviews timed to your own forgetting.

26 lessons in Machine Learning: Doing It Honestly

Start from a baselineFour counts, not one numberPrecision, recall and the thresholdCount people, not percentagesThe same model, ten runsHow much the test set can tell youComparing two models fairlyAblation: take one part outLearning pure noiseCopies on both sides of the splitWhen the future leaks into the pastPeeking at the test setTrain, validation, testLeakage in published researchWhen the labels are wrongData from somewhere elseWrite down where the data came fromFair to whom?You cannot equalise them allReproduce a claim, including oursReport it so others can checkA model cardThree beliefs that surviveLearning forever, finding its own data, understandingWhen not to use machine learningAn honest experiment, start to finish

Built to be understood — and remembered.

Every idea is taught with motivation and a worked example before the drills, and an FSRS spaced-repetition engine schedules each review for the day its model predicts you would forget it. A short placement check finds what you already know, so you start Machine Learning: Doing It Honestly exactly where it's useful.

Start Machine Learning: Doing It Honestly →
Related Ml subjects
Machine Learning: The Whole GameMachine Learning: Learning from Data, in CodeMachine Learning: Neural Networks from ScratchMachine Learning: Build Your Own Language ModelMachine Learning: Deep Learning, Seeing and Remembering