subtopic

Loss function: measuring the mistake

10 min applied40 min research1 curated resources

Learn that the loss is one number saying how wrong the network is. Use cross-entropy for classification and MSE/MAE for regression (numbers).

Builds on: B17 · Build: Fit a Line with Gradient Descent (calculus in action), B18 · Build: Spam Classifier with Probabilities (sigmoid, softmax, cross-entropy), S01 · Build: How Training Works (fit, loss, optimisation)

Notes, key code, what to skip and 1 curated resources are inside the platform.Join the waitlist and we'll build you a roadmap that includes this topic only if your goal needs it.

Get my roadmap