Skip to content
Discussion options

You must be logged in to vote

JSON Lines files for v0.1, Postgres once the API is serving.

Right now every result comes from offline runs over a fixed test set. JSONL (one InspectionResult per line, one file per experiment run) is append-only, diffable, trivially reproducible, and the evaluation harness can load it straight into a dataframe. A database adds migrations and a running service for no gain yet.

The switch point is when the API handles live uploads: concurrent writes, and queries like "all anomalous parts from line 3 this week". At that point a Postgres table with the scalar fields as columns and defects as JSONB fits well.

Because everything goes through the Pydantic model, the storage layer is one small i…

Replies: 1 comment

Comment options

You must be logged in to vote
0 replies
Answer selected by D-M-Pearman
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Category
Q&A
Labels
None yet
2 participants