lesson 5 of 6 · continuous training and rolloutlesson 5 of 6 · continuous training and rolloutlesson 5 of 6 · continuous training and rolloutlesson 5 of 6 · continuous training and rolloutlesson 5 of 6 · continuous training and rolloutlesson 5 of 6 · continuous training and rolloutlesson 5 of 6 · continuous training and rolloutlesson 5 of 6 · continuous training and rolloutlesson 5 of 6 · continuous training and rolloutlesson 5 of 6 · continuous training and rolloutlesson 5 of 6 · continuous training and rollout
Checkpoint
One last thing before we move on. pass this to mark the lesson done, or skip and keep moving. hop to the next when you're ready.
Write evaluate(canary, incumbent) — the rollout decision as code. Print the comparison line f"canary {c:.1%} (n={cn}) vs incumbent {i:.1%} (n={inn})", then the verdict: "-> SOAK: not enough canary traffic for a verdict" below MIN_N=500; "-> PROMOTE" at-or-above the incumbent; otherwise "-> ROLLBACK: registry promote(previous), one command, no meeting". Run the two readings: an early n=200 look, then the soaked n=1000 one.