2 Commits

Include renames

Author	SHA1	Message	Date
mathiasandClaude Sonnet 4.6	20aeecb971	fix(eval): correct probe metric to use true year-based OOS split ... - train.py build(): year-based split (train≤2021, OOS≥2022) replaces misleading 70/30 mixed-period split; true OOS val_vol_r2 now ~-0.36 vs previously reported +0.18 (artefact of cross-period data leakage) - train.py: EXPORT_EMBEDDINGS block now exports both train+OOS embeddings with dates and HV labels for Go eval harness - cmd/eval: LinearProbeTrainTest uses train stats for standardisation of both sets (no leakage); standardiseCompute/applyStandardise helpers - internal/eval: add LinearProbeTrainTest (fit-on-train, eval-on-OOS) alongside LinearProbe (same-set); 8/8 tests still green Phase-0 gate result: val_vol_r2=-0.36, silhouette=0.043, erank=58.9/64. Backbone produces high-rank embeddings (SIGReg working) but does NOT generalize across 2021→2022 regime boundary. Gate: INCONCLUSIVE/FAIL. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>	2026-06-24 22:57:14 +02:00
mathiasandClaude Sonnet 4.6	e11e7d2524	feat(eval): Go evaluation harness — LinearProbe, Silhouette, EffectiveRank (#4) ... internal/eval: three pure-Go diagnostics on frozen embeddings: LinearProbe(emb, y, λ) → val_vol_r2 (OOS R², closed-form ridge, Cholesky) Silhouette(emb, labels) → mean silhouette (Euclidean, multi-label, errors on <2 classes) EffectiveRank(emb) → Roy effective rank (Jacobi eigenvalues → entropy → exp(H)) cmd/eval/main.go: CLI driver reading embeddings.json (exported by train.py with EXPORT_EMBEDDINGS=1), standardises per-dim, dispatches to -metric flag. task eval:probe / eval:silhouette / eval:collapse wired in Taskfile. 8/8 tests pass (red-green: perfect clusters, rank-1, full-rank, noise, constant target, single-label error). Pure stdlib, no external deps. Closes #4. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>	2026-06-24 12:01:04 +02:00