berkus
/
wifi-densepose
mirror of https://github.com/ruvnet/wifi-densepose.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity
893 Commits 375 Branches 133 Tags 193 MiB
Commit Graph

4 Commits

Author SHA1 Message Date
ruv 96ccfa58fb bench: ship int4 edge artifact + CPU latency 
Published deployable int4-QAT micro (verified 74.08%, ~20KB) at
ruvnet/wifi-densepose-mmfi-pose/edge. Runs 0.135ms single-thread x86 CPU
(no GPU) - real-time pose without an accelerator. ARM on-device validation
pending fleet availability.

Co-Authored-By: claude-flow <ruv@ruv.net>
 2026-05-31 01:30:29 -04:00
ruv 92d433523d bench: deployed quantized accuracy + QAT for micro edge model 
int8 PTQ lossless (74.70%, 73.5KB); int4 naive PTQ drops below SOTA
(70.21%) but QAT recovers to 74.46% (36.7KB) - still beats MultiFormer.
A SOTA-beating WiFi-pose model genuinely runs in ~37KB int4 (QAT) /
73KB int8. Distillation negative noted.

Co-Authored-By: claude-flow <ruv@ruv.net>
 2026-05-31 01:23:30 -04:00
ruv d64323c2d6 bench: add quantized footprint — SOTA-beating WiFi pose in 37KB int4 
micro (74.87%, beats MultiFormer 72.25%) = 36.7KB int4 / 73.5KB int8;
nano (~72%) = 19.5KB int4. Distillation tested, no gain (direct training
wins). A SOTA-beating pose model fits on the sensing node itself.

Co-Authored-By: claude-flow <ruv@ruv.net>
 2026-05-31 01:16:16 -04:00
ruv 9c64d90054 bench: WiFi-CSI pose efficiency frontier — 75K-param model beats SOTA 
Swept model size on MM-Fi random_split: every config from micro (75,237
params, 0.22ms, 74.30%) up beats MultiFormer (72.25%); nano (40K, 0.13ms)
within 0.5pt. Pareto-dominant (smaller AND more accurate than prior SOTA).
Orthogonal to the data-bound accuracy frontier (ADR-150).

Co-Authored-By: claude-flow <ruv@ruv.net>
 2026-05-31 01:10:33 -04:00