EVAL.md

eval.py: Floor sanity check

A minimal smoke test. It confirms your predictions parse and have the right shape. It does NOT score recommendation quality.

For deeper scoring (keyword matching, multi-tier reasoning, fixture citations), bring your own evaluator that compares each prediction against the handcrafted_recommendation.json in the matching scenario folder.

Usage

python eval.py --predictions sample_predictions.json

What it checks

1. The file parses as JSON.
2. There is a top-level predictions array.
3. Each prediction has the required fields.
4. Each finding_type is one of the three allowed values.
5. Each primary_tier is one of the allowed tier names (or null).
6. Each action_category is one of the allowed values.
7. Each specific_change is at least 20 characters.

What it does NOT check

• Whether the recommendation engages with the right evidence.
• Whether multi-tier scenarios cite both tiers.
• Whether named fixtures from the metadata are referenced.
• Whether cost or projection numbers are reasonable.

Those checks are quality assessments. They depend on what you want to score for and how strict you want to be. The dataset ships the gold answers; the scoring method is up to you.

Exit codes

• 0: all predictions passed the Floor check.
• 1: at least one prediction failed at least one check.
• 2: usage error (missing file, malformed JSON).