From b5f52aeeb173b94523edaebbfd940a945bdadd59 Mon Sep 17 00:00:00 2001 From: Simba Zhang Date: Sat, 7 Mar 2026 08:54:31 -0800 Subject: [PATCH 1/2] fix: increase benchmark ping max_completion_tokens from 1 to 5 A single token is often insufficient for the model to produce a valid response token, causing false ping failures. --- .../analysis/home-security-benchmark/scripts/run-benchmark.cjs | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/skills/analysis/home-security-benchmark/scripts/run-benchmark.cjs b/skills/analysis/home-security-benchmark/scripts/run-benchmark.cjs index cbac10eb..b45306cd 100644 --- a/skills/analysis/home-security-benchmark/scripts/run-benchmark.cjs +++ b/skills/analysis/home-security-benchmark/scripts/run-benchmark.cjs @@ -1775,7 +1775,7 @@ async function main() { const ping = await llmClient.chat.completions.create({ ...(LLM_MODEL && { model: LLM_MODEL }), messages: [{ role: 'user', content: 'ping' }], - max_completion_tokens: 1, + max_completion_tokens: 5, }); results.model.name = ping.model || 'unknown'; log(` Model: ${results.model.name}`); From 316aadcf1795cde5814844007594eb68273e2577 Mon Sep 17 00:00:00 2001 From: Simba Zhang Date: Sat, 7 Mar 2026 09:24:12 -0800 Subject: [PATCH 2/2] =?UTF-8?q?fix:=20correct=20broken=20source=20code=20l?= =?UTF-8?q?ink=20(detector=5Fcpu.py=20=E2=86=92=20detector.py)?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit --- docs/legacy-applications.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/docs/legacy-applications.md b/docs/legacy-applications.md index 6db48503..cc0a82fd 100644 --- a/docs/legacy-applications.md +++ b/docs/legacy-applications.md @@ -7,7 +7,7 @@ ## Application 1: Self-supervised Person Recognition (REID) for Intruder Detection -SharpAI yolov7_reid is an open source python application that leverages AI technologies to detect intruders with traditional surveillance cameras. [Source code](https://github.com/SharpAI/DeepCamera/blob/master/src/yolov7_reid/src/detector_cpu.py) +SharpAI yolov7_reid is an open source python application that leverages AI technologies to detect intruders with traditional surveillance cameras. [Source code](https://github.com/SharpAI/DeepCamera/blob/master/src/yolov7_reid/src/detector.py) It leverages Yolov7 as person detector, FastReID for person feature extraction, Milvus the local vector database for self-supervised learning to identify unseen persons, Labelstudio to host images locally and for further usage such as labeling data and training your own classifier. It also integrates with Home-Assistant to empower smart home with AI technology.