Diagnose three learning curves and choose a fitting action instead of adding layers at random.
Proof of success: explain "Underfitting and overfitting", complete "Curve Doctor" and name the limit of the analogy.
Mila trains three models on symbols. The first is wrong a lot on both train and validation: it isn't learning enough. The second succeeds on training but gets worse on validation: it memorizes details. The third improves both and then stabilizes with a reasonable gap. Mila doesn't pick the model with the best training score; she reads both curves, inspects errors and checks the result across several batches.
Preview — the rest of the lesson is for enrolled readers.
Your access is tied to your account, not to this link. Sign in with the same email you used in class: your course is waiting, no need to enter the code again.
Sign inNo account yet? Create oneThe first modules of the course are open to everyone. For the rest you have three options: buy this course once and for all, subscribe, or enter the code handed out in class.