liveBook · Manning

Accuracy

As explained in Machine Learning System Design

Accuracy is the proportion of evaluated predictions that match the corresponding correct labels.

Prediction accuracy is a monitoring signal: compare batch predictions with actual demand to assess whether they are producing high-quality results. Track it alongside job success, duration, failures, and data validation checks. section 15.5.1

Accuracy matters differently depending on the cost of a mistake. An entertainment effect that looks wrong for a frame may be tolerable; a medical or transport system should favor returning no result over an inaccurate one. section 2.4

How Machine Learning System Design covers it

How deep it goes
Names prediction accuracy as a monitoring metric but does not define its calculation.
Examples
Contrasts an ugly AR-mask frame with inaccurate medical or transport results, and tracks batch predictions against actual demand.
What you can do after
Evaluate batch predictions by tracking accuracy against actual demand.
How it connects
Connects accuracy to correctness, robustness, and batch-job monitoring.

Also explained in

Compare how 32 books explain accuracy →

Related concepts

This explainer was generated by Manning's knowledge graph from the text of Machine Learning System Design. The section links point to the parts of the book it draws on.