A historical-society minute-book project prints a final number but never checks whether validation tracked training or diverged after epoch two. What should they read as the primary experiment instrument?
Select an answer to reveal the explanation.
Short Explanation
A final number without the train-versus-validation pair hides overfit, underfit, or a late collapse. Read those two curves as the primary instrument. Fan speed, the first loss value, and a multimodal board are wallpaper.
Full Explanation
Train and validation curves are the instrument of the run, not wallpaper next to a final number. When the two tracks stay together, the fit is still useful; when validation diverges after an early epoch, the run has started to overfit or collapse. A single end-of-run score hides that timing. Fan speed, the first loss value, and multimodal system boards are not that instrument.