The glossary
Every word you'll meet in AI and machine learning, explained in everyday language.
Software that records each step of a calculation, then walks back through the record to find the slope of the result with respect to every input.
The tape ends empty, as PyTorch's does after backward(). To walk it again, you run the forward pass again, which records a fresh tape.
Automatic differentiation: how PyTorch finds every slope · A tape of steps walked backward, and requires_grad, backward(), .grad, zero_grad() and no_grad(). About 40 minutes.
1286 terms in the glossary, each with the lesson that teaches it.
Every word you'll meet in AI and machine learning, explained in everyday language.
A tape of steps walked backward, and requires_grad, backward(), .grad, zero_grad() and no_grad().
Never trained an AI? Teach a small one to write, right in your browser. It takes about 20 seconds and needs no account. Teach it to write