mathias
308f71b566
fix(ci): Taskfile YAML syntax error + staticcheck tagged-switch (jepa-fx-risk#19)
...
CD / Build & Import (push) Failing after 10s
CD / Deploy via GitOps (push) Has been skipped
CD / Lint / Test / Vet (push) Successful in 7s
Taskfile.yml line 46 had an unquoted Go-template `{{.VAR}}` inside a YAML
flow sequence (`cmds: [...]`) -- the literal braces broke YAML parsing
outright ("did not find expected ',' or ']'"), so `task check` (and thus
CI's push-triggered check job) failed before running a single command.
A manual `workflow_dispatch` re-run passed because the autoresearch
workflow never calls `task check` at all -- unrelated path, not an
env/secret difference as first suspected. Quoted the string.
Also fixed a staticcheck QF1002 in internal/eval/var.go: a boolean
switch comparing the same variable (n1) in every case is a tagged
switch in disguise -- converted to `switch n1 { case 0: ... case n: ...
}`.
2026-07-24 15:26:01 +02:00
mathias and Claude Sonnet 4.6
d4b67943fb
feat(multipair): lock best config + train:multipair Taskfile target
...
CD / Lint / Test / Vet (push) Failing after 2s
CD / Build & Import (push) Has been skipped
CD / Deploy via GitOps (push) Has been skipped
Best: USE_MULTIPAIR=1 D_MODEL=256 WINDOW=120 → phase1_r2=0.4377, val_vol_r2=0.3695
WINDOW sweep (multipair D=256): W=60→0.4009, W=120→0.4377, W=240→0.4277
D_MODEL=128 undercapacity for 10ch (val_vol_r2=0.1013); D=256 restores probe quality.
Single-pair default knobs unchanged (D=128 still optimal there).
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com >
2026-06-26 14:08:25 +02:00
mathias and Claude Sonnet 4.6
bd8962f997
feat(multipair): G10 multi-pair pipeline + USE_MULTIPAIR knob (Option C)
...
- prepare_hourly.py: parameterize PAIR env var; OUT_DEFAULT per-pair; load_m1_from_zips(pair=)
- prepare_multipair.py: inner-join 5-pair hourly parquets on datetime → wide parquet
cols: datetime, {pair}_ret, {pair}_rv × n_pairs; eurusd_rv = target
- fetch_multipair.py: download GBPUSD/USDJPY/USDCHF/AUDUSD M1 2008-2023 from histdata
- train.py: USE_MULTIPAIR knob (JEPA_USE_MULTIPAIR=1); build() reads multipair parquet
with n_channels = n_pairs × 2; target = eurusd_rv
- Taskfile: data:fetch:multipair, data:prepare:pair, data:prepare:multipair, data:test updated
- 7 new tests in test_multipair.py; 34/35 pass (1 SKIP until multipair parquet built)
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com >
2026-06-26 14:00:57 +02:00
mathias and Claude Sonnet 4.6
e31905dc43
feat(data): EUR/USD hourly pipeline + 2008-2023 M1 dataset ( #2 )
...
CD / Lint / Test / Vet (push) Successful in 4s
CD / Build & Import (push) Failing after 8s
CD / Deploy via GitOps (push) Has been skipped
- scripts/prepare_hourly.py: M1→hourly aggregation (realized_vol = sqrt(Σr²),
MIN_BARS=30 threshold, no weekend rows, year-based split preserved)
- tests/test_prepare_hourly.py: 5 TDD tests, all green
- train.py: USE_HOURLY=True, WINDOW=240 (10-day), PATCH_LEN=24 (1-day patches);
build() prefers eurusd_hourly.parquet, falls back to daily; EXPORT BLOCK updated
- Taskfile.yml: data:fetch:historical, data:prepare:hourly, data:prepare:all, data:test
- 98,591 hourly rows (2008-2023) covering GFC, Euro crisis, Brexit, COVID, Fed cycle
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com >
2026-06-25 13:12:48 +02:00
mathias and Claude Sonnet 4.6
e11e7d2524
feat(eval): Go evaluation harness — LinearProbe, Silhouette, EffectiveRank ( #4 )
...
CD / Build & Import (push) Failing after 7s
CD / Deploy via GitOps (push) Has been skipped
CD / Lint / Test / Vet (push) Successful in 4s
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
mathias
d1193a57c6
Initial commit
2026-05-27 21:55:18 +00:00
mathias
