Instrument the model-run dictionary

Instrument the model-run record dictionary so a run records enough structure to debug offline: counts, latency if relevant, pass/fail of assert required keys, and a stable stage name. Redact secrets and raw credentials from every event.

import json
print(json.dumps({'event':'run_record','keys':['name','accuracy','ready','owner','version'],'ready':True}))

Expected evidence: JSON event listing keys. Prefer JSON or structured text you can grep in CI over prose logs for python-dictionaries.

Turn one historical failure—especially KeyError on missing version—into a test that fails if the signal disappears for the model-run dictionary. Observability without a failing test is optional decoration; observability with a test is part of the python-dictionaries artifact.

Draft a three-field event for the model-run dictionary: stage, ok, and one domain field derived from assert required keys; .get('version','unversioned') returns a string. Add fixture_id or docs_version when content can change. Explicitly list fields that must never appear (tokens, passwords, raw prompts) because serializing secrets (API keys) into the run record JSON is in scope for this lab.

Wire one assertion that fails if the model-run dictionary event is missing after a run. Observability that cannot fail a test will not survive contact with a busy python-dictionaries repository.

Imagine a teammate opens only your event stream after a bad deploy. Could they tell whether name='ready-v1', accuracy=0.91, ready=True, owner='ml-lab' was wrong, whether KeyError on missing version, or silent overwrite of accuracy with a string returned, or whether serializing secrets (API keys) into the run record JSON slipped through? If not, rename fields until those three stories are distinguishable.

At the testing and observability stage for python-dictionaries, the job is narrower than finishing a product demo. You are creating one progressive evidence piece about name='ready-v1', accuracy=0.91, ready=True, owner='ml-lab' 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: a handwritten field list before coding the dict.

For this page specifically, success looks like a structured event schema locked by a test while still centering the user decision to store one training run so later code can read named fields without positional guessing. If you cannot point to a file, command, or assertion that proves that for the model-run dictionary, stay on this page instead of advancing.

ML Python starter

Previous · Next