26 lessons in Machine Learning: Neural Networks from Scratch

Where one neuron failsA layer of neuronsWhy the squash mattersA layer in NumPyHow many numbers in a networkRates multiplyA calculation as a graphLocal slopesUsed twice, counted twiceA value that remembersMethods, and a readable printa * b + c builds a graphEach operation knows its way backThe order to visitEvery slope in one callA neuron the engine can trainLayers and networks of ValuesTrain it, and the bug that hidesEvery point at onceTrain it and draw itWhere the weights startReading the loss curveMomentum and AdamA few at a timeThree species: softmaxWhen training goes wrong

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: Neural Networks from Scratch exactly where it's useful.

Start Machine Learning: Neural Networks from Scratch →

Related Ml subjects

Machine Learning: The Whole Game Machine Learning: Learning from Data, in Code Machine Learning: Build Your Own Language Model Machine Learning: Deep Learning, Seeing and Remembering Machine Learning: Doing It Honestly