Skip to content

v0.3.0

Choose a tag to compare

View all tags
@mmcdermott mmcdermott released this 21 Apr 02:56
· 32 commits to main since this release
v0.3.0
e7e637e
This commit was created on GitHub.com and signed with GitHub’s verified signature.
GPG key ID: B5690EEEBB952194
Verified
Learn about vigilant mode.

Phase 1 + Phase 2.1-2.5 refactor roll-up (merged via #131).

New CLIs: EQ_generate_training_tasks (renamed from EQ_generate_tasks), EQ_generate_evaluation_tasks, EQ_predict.

EQ_evaluate rewire: consumes PredictionSchema parquet → per-(query, duration_days) metrics. Legacy four-stage evaluator (eval.py / gen_index_times / gen_task / select_model) deleted.

Cross-stage TaskQuerySchema: single source-of-truth schema across generate_tasks → predict → evaluate, with nullable boolean_value as the censored-row sentinel.

Other behavior changes: EQ_predict now hard-raises on out-of-vocab query codes (was: warn), refuses to overwrite an existing output_parquet without overwrite=true, and asserts SequentialSampler at startup.

See #131 for the full description + caveats.

Assets 2
Loading