| Add wall-clock checkpoints and full eval defaults
| 2 个月前 |
| Merge origin/main into feat/ast-aware-chunking
Resolve conflicts: combine AST chunking args (filepath, chunkStrategy)
with abort signal parameter from #458.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
| 1 个月前 |
| lots of training stuff
| 3 个月前 |
| lots of training stuff
| 3 个月前 |
| lots of training stuff
| 3 个月前 |
| lots of training stuff
| 3 个月前 |
| finetune: quoted phrases, negation, and entity preservation (#247)
Training data:
- Expand lex phrases/negation examples from 12 to 74 with intent field
- Add 50 personal entity examples (meetings, emails, projects with names)
Reward function:
- Detect entities at position 0 (fixes "Bob asked about deploy")
- Per-entity coverage penalty: -20 per entity absent from all lex+vec
- Phrase quoting bonus: +3 when lex uses quotes for multi-word terms
- Expanded stopwords to reduce false positive entity detection
Eval queries: add 21 test queries for personal entities, quoted phrases,
and negation/disambiguation scenarios.
| 2 个月前 |
| lots of training stuff
| 3 个月前 |
| lots of training stuff
| 3 个月前 |
| finetune: quoted phrases, negation, and entity preservation (#247)
Training data:
- Expand lex phrases/negation examples from 12 to 74 with intent field
- Add 50 personal entity examples (meetings, emails, projects with names)
Reward function:
- Detect entities at position 0 (fixes "Bob asked about deploy")
- Per-entity coverage penalty: -20 per entity absent from all lex+vec
- Phrase quoting bonus: +3 when lex uses quotes for multi-word terms
- Expanded stopwords to reduce false positive entity detection
Eval queries: add 21 test queries for personal entities, quoted phrases,
and negation/disambiguation scenarios.
| 2 个月前 |
| lots of training stuff
| 3 个月前 |
| data: add 48 sports acronym training examples
Covers UFC, NFL, NBA, NHL, MLB, F1, MLS, IMSA, WEC, NASCAR, PGA, ATP, WTA, FIFA.
Fixes query expansion failures like UFC → 'united fighting club'.
| 2 个月前 |
| finetune: strict Pydantic schema, one canonical data format
Replace ad-hoc JSON parsing with a strict Pydantic model
(TrainingExample with typed OutputPair). All data loading goes
through load_examples() which fails loudly on invalid data.
- Convert v3_structured.jsonl from "searches" to "output" format
- Rewrite all consumer scripts (prepare, validate, score, analyze)
to load through the Pydantic schema
- Prepared train/val files are ephemeral build artifacts
- Restore LFM2 and GEPA experiments under experiments/
- Add pydantic>=2.0 to dependencies
| 2 个月前 |
| lots of training stuff
| 3 个月前 |