ruv 946acf2d10 docs(cog-pose): correct misleading adapter cross-reference ...

The --adapter docs claimed the adapter is produced by
aether-arena/calibration/calibrate.py, but that reference tool targets the
MM-Fi *transformer* model and emits .npz with proj/head LoRA keys, while
this cog runs a *conv+MLP* model expecting safetensors with fc1.a/fc1.b/
fc2.a/fc2.b. Same LoRA mechanism, different model -> adapters are
model-specific and NOT interchangeable. Clarify the expected key layout and
that the Python tool is a mechanism reference, not a drop-in producer.
6/6 tests pass.

Co-Authored-By: claude-flow <ruv@ruv.net>

2026-05-31 05:04:35 -04:00

..

cog

docs: repoint #640 references to #645 (original deleted, replaced) (#646)

2026-05-19 17:18:05 -04:00

src

docs(cog-pose): correct misleading adapter cross-reference

2026-05-31 05:04:35 -04:00

tests

feat(cog-pose): per-room LoRA calibration adapter in the Rust inference path

2026-05-31 02:26:48 -04:00

Cargo.toml

chore(cogs): publish cog-person-count + cog-pose-estimation 0.3.0 to crates.io

2026-05-25 10:52:47 -04:00