Product recognition demo for supermarket shelf images

kiyo9w 30c0292ed1 perf: keep CPU for detection, then document the measured defaults ... Detection and classification share the CPU, and leaving both uncapped made them fight: uncapped torch plus 8 workers measured 11.1 results/s single-client and dropped to 6.3/s each with two clients. Capping torch at 4 threads and classifying with 5 workers gives 12.2 results/s single-client (46 ms detection) and 8.5/s each for two clients, with zero unexplained label changes. Detection is the part that must stay real-time; classification fills in behind it, so the defaults now protect detection's share of the CPU. Both remain overridable (--classify-workers, DORAYAKI_TORCH_THREADS). docs/API.md records the shipped numbers and says which configuration produced them, including the honest gap: detection during a live pan is ~46-48 ms, not the 28 ms of an idle measurement. Co-Authored-By: Devin <158243242+devin-ai-integration[bot]@users.noreply.github.com>		2026-09-24 13:36:41 +09:00
recognization	perf: keep CPU for detection, then document the measured defaults	2026-09-24 13:36:41 +09:00
.gitignore	Initial commit for push folder	2026-09-18 09:13:53 +09:00