Debug training on the test row in the tiny linear model

Do not start with a speculative fix for the tiny linear model. Force the failure on purpose, save the before output, then change one cause at a time. Retries are allowed only for transient conditions—not for bad input that will fail forever on train-a-tiny-model.

all_X=[1,2,3,4,5]; print({'leak_if_fit_on':all_X,'fix':'fit only on train hours'})

Expected evidence: **leak warning**. If you cannot reproduce on demand, you do not yet control the failure mode for `train-a-tiny-model`.

After repair, rerun the exact reproduction command. Keep the failing fixture as a regression seed for the observability page. For the hours→score linear fit with holdout, remember the claim you are restoring: fitted slope/intercept, one held-out prediction, and MAE are all printed.

Store a one-command reproduction for: training on the test row, or reporting R² without the holdout error. The command should use small hours/score table with one held-out pair or a minimal mutant of it. Paste the failing output into notes/failure-before.txt (or your shell scrollback as copied text). After the fix, paste notes/failure-after.txt and keep both.

Retries belong only on transient faults. If the failure is bad input, a bad allowlist, or a logic bug in the tiny linear model, retrying will amplify cost without repairing trust around estimate slope/intercept from study hours and score a held-out row honestly.

Classify the failure as prevent, detect, contain, or recover—using this lab’s language, not a generic poster. For train-a-tiny-model, the first fix should usually be detect+prevent at the boundary, because training on the test row, or reporting R² without the holdout error is cheaper to stop early than to explain in production prose.

At the debugging stage for train-a-tiny-model, the job is narrower than finishing a product demo. You are creating one progressive evidence piece about small hours/score table with one held-out pair that later pages inherit without redefining success. Keep that fixture small enough to inspect by hand, keep outputs copy-pasteable as text, and refuse to narrate this baseline as if it were a production SLA: predict mean train score for the holdout before fitting.

For this page specifically, success looks like before/after evidence for the characteristic failure while still centering the user decision to estimate slope/intercept from study hours and score a held-out row honestly. If you cannot point to a file, command, or assertion that proves that for the tiny linear model, stay on this page instead of advancing.

Overfitting glossary

Previous · Next