mathias/jepa-fx-risk

Fork 0

generated from mathias/template-go-web

T

mathiasandClaude Sonnet 4.6 20aeecb971

CD / Lint / Test / Vet (push) Successful in 4s

Details

CD / Build & Import (push) Failing after 7s

Details

CD / Deploy via GitOps (push) Has been skipped

Details

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

.context

docs: add DECISIONS.md and rewrite PROJECT.md (#8 )

2026-06-22 18:10:32 +00:00

.gitea/workflows

Initial commit

2026-05-27 21:55:18 +00:00

.skills

Initial commit

2026-05-27 21:55:18 +00:00

cmd

fix(eval): correct probe metric to use true year-based OOS split

2026-06-24 22:57:14 +02:00

internal

fix(eval): correct probe metric to use true year-based OOS split

2026-06-24 22:57:14 +02:00

results/summaries

experiment(phase0): NULL result — path B proxy gate (ref #5 )

2026-06-24 10:52:07 +02:00

scripts

experiment(phase0): NULL result — path B proxy gate (ref #5 )

2026-06-24 10:52:07 +02:00

.aider.conf.yml

Initial commit

2026-05-27 21:55:18 +00:00

.aider.conventions.md

Initial commit

2026-05-27 21:55:18 +00:00

.cursorrules

Initial commit

2026-05-27 21:55:18 +00:00

.gitignore

feat(data): EUR/USD M1 fetch + daily realized-vol prep (toy slice, #2/#11)

2026-06-23 16:49:59 +02:00

AGENTS.md

Initial commit

2026-05-27 21:55:18 +00:00

CLAUDE.md

Initial commit

2026-05-27 21:55:18 +00:00

DECISIONS.md

docs: add DECISIONS.md and rewrite PROJECT.md (#8 )

2026-06-22 18:10:32 +00:00

Dockerfile

Initial commit

2026-05-27 21:55:18 +00:00

go.mod

refactor: rename hostexecutor → jepa-fx-risk (#9 )

2026-06-24 11:58:42 +02:00

go.sum

refactor: rename hostexecutor → jepa-fx-risk (#9 )

2026-06-24 11:58:42 +02:00

loop.py

feat(loop): autoresearch keep/revert loop + first iteration (val_vol_r2 0.2821→0.3749, +9.3%)

2026-06-24 07:16:50 +02:00

metrics.json

feat(loop): 5 iters on TS-JEPA+SIGReg backbone — consistent improvement

2026-06-24 07:45:55 +02:00

README.md

Initial commit

2026-05-27 21:55:18 +00:00

requirements.txt

feat(data): EUR/USD M1 fetch + daily realized-vol prep (toy slice, #2/#11)

2026-06-23 16:49:59 +02:00

STATUS.md

feat(loop): 5 iters on TS-JEPA+SIGReg backbone — consistent improvement

2026-06-24 07:45:55 +02:00

Taskfile.yml

feat(eval): Go evaluation harness — LinearProbe, Silhouette, EffectiveRank (#4 )

2026-06-24 12:01:04 +02:00

train.py

fix(eval): correct probe metric to use true year-based OOS split

2026-06-24 22:57:14 +02:00

README.md

hostexecutor

Generated from mathias/template-go-web.

Bootstrap

After creating from template, run:

go mod tidy       # regenerate go.sum with real module path
task generate     # generate templ files
task build        # build the binary