[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M753H42XR0NMCSZPZ22DS5DA":3,"skill-same:01M753H42XR0NMCSZPZ22DS5DA":66},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":15,"categories":16,"keywords":17,"publishedAt":18,"linkOnly":19,"markdownContent":20,"markdownBody":21,"frontMatter":22,"refName":26,"commitSha":27,"skillDirectory":28,"forksCount":29,"repositoryHtmlUrl":30,"repositoryHomepageUrl":6,"ownerHtmlUrl":31,"ownerBlogUrl":6,"ownerDescription":6,"updatedAt":32,"install":33,"resources":45},"01M753H42XR0NMCSZPZ22DS5DA","onnx-and-benchmarking",null,"Use 3DDFA_V2 ONNX Runtime acceleration and CPU latency or speed benchmark workflows.","VectorSpaceLab\u002FAREX-Skill","VectorSpaceLab","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F180637102?v=4","01M7538C5WPSJGWF7KHR4EFZ9F","01M7538C6R2GFP53NPXTCZNBRT",351,"MIT","https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002F3ddfa-v2\u002Fsub-skills\u002Fonnx-and-benchmarking\u002FSKILL.md",[],[],"2026-11-12 13:55:43",false,"---\nname: \"onnx-and-benchmarking\"\ndescription: \"Use 3DDFA_V2 ONNX Runtime acceleration and CPU latency or speed\n  benchmark workflows.\"\nmetadata:\n  disco-role: operating\ndisable-model-invocation: true\nlicense: MIT\n---\n\n# ONNX and benchmarking\n\nUse this sub-skill when the task is about `--onnx`, ONNX conversion, CPU\nlatency, `OMP_NUM_THREADS`, `latency.py`, or `speed_cpu.py`.\n\n## When to read\n\nRead this sub-skill when the task asks to:\n\n- Use ONNX Runtime instead of the PyTorch model path.\n- Explain or prepare `TDDFA_ONNX` and `FaceBoxes_ONNX`.\n- Run end-to-end latency timing on an image.\n- Run the model-only CPU microbenchmark.\n- Diagnose missing `.onnx` files, OpenMP\u002Flibomp issues, or unstable timings.\n\n## Before running\n\n1. Use `..\u002Fsetup-and-assets\u002F` to build and verify the baseline checkout.\n2. Check `..\u002F..\u002Freferences\u002Fmodel-assets.md` for which `.pth`, `.pkl`, and\n   generated `.onnx` assets are involved.\n3. Keep benchmark inputs and thread counts fixed when comparing numbers.\n\n## Latency wrapper\n\nThe latency wrapper preserves the original `latency.py` CLI. Put original\narguments after `--`:\n\n```bash\npython \u003Cskill-root>\u002Fsub-skills\u002Fonnx-and-benchmarking\u002Fscripts\u002Frun-latency.py \\\n  --repo-root \u003Ccheckout> -- \\\n  -f \u003Cimage-path> --onnx --repeated 1 --warmup true\n```\n\nThis reports face detection, 3DMM regression, and reconstruction timing.\n\n## Speed wrapper\n\nThe speed wrapper preserves `speed_cpu.py`:\n\n```bash\npython \u003Cskill-root>\u002Fsub-skills\u002Fonnx-and-benchmarking\u002Fscripts\u002Frun-speed-cpu.py \\\n  --repo-root \u003Ccheckout>\n```\n\nIt runs a random input through `weights\u002Fmb1_120x120.onnx` and prints the mean\nand standard deviation per inference.\n\n## ONNX conversion behavior\n\n- `FaceBoxes_ONNX` converts `FaceBoxesProd.pth` to `FaceBoxesProd.onnx` if the\n  ONNX file is missing.\n- `TDDFA_ONNX` converts the selected `.pth` checkpoint and the BFM decoder if\n  the ONNX artifacts are missing.\n- Conversion is CPU-safe but mutates the checkout by writing generated `.onnx`\n  files.\n\n## Thread and provider notes\n\n- The repo sets `OMP_NUM_THREADS=4` in the ONNX demo paths.\n- `KMP_DUPLICATE_LIB_OK=True` is set by the demo scripts before ONNX imports.\n- CPU ONNX Runtime is the baseline. CUDA\u002Fother providers are outside the\n  required verification scope.\n- On macOS, the original README notes that `libomp` may be required.\n\n## Troubleshooting\n\nRead `references\u002Ftroubleshooting.md` for ONNX-specific issues and\n`..\u002F..\u002Freferences\u002Ftroubleshooting.md` for shared setup failures.\n","# ONNX and benchmarking\n\nUse this sub-skill when the task is about `--onnx`, ONNX conversion, CPU\nlatency, `OMP_NUM_THREADS`, `latency.py`, or `speed_cpu.py`.\n\n## When to read\n\nRead this sub-skill when the task asks to:\n\n- Use ONNX Runtime instead of the PyTorch model path.\n- Explain or prepare `TDDFA_ONNX` and `FaceBoxes_ONNX`.\n- Run end-to-end latency timing on an image.\n- Run the model-only CPU microbenchmark.\n- Diagnose missing `.onnx` files, OpenMP\u002Flibomp issues, or unstable timings.\n\n## Before running\n\n1. Use `..\u002Fsetup-and-assets\u002F` to build and verify the baseline checkout.\n2. Check `..\u002F..\u002Freferences\u002Fmodel-assets.md` for which `.pth`, `.pkl`, and\n   generated `.onnx` assets are involved.\n3. Keep benchmark inputs and thread counts fixed when comparing numbers.\n\n## Latency wrapper\n\nThe latency wrapper preserves the original `latency.py` CLI. Put original\narguments after `--`:\n\n```bash\npython \u003Cskill-root>\u002Fsub-skills\u002Fonnx-and-benchmarking\u002Fscripts\u002Frun-latency.py \\\n  --repo-root \u003Ccheckout> -- \\\n  -f \u003Cimage-path> --onnx --repeated 1 --warmup true\n```\n\nThis reports face detection, 3DMM regression, and reconstruction timing.\n\n## Speed wrapper\n\nThe speed wrapper preserves `speed_cpu.py`:\n\n```bash\npython \u003Cskill-root>\u002Fsub-skills\u002Fonnx-and-benchmarking\u002Fscripts\u002Frun-speed-cpu.py \\\n  --repo-root \u003Ccheckout>\n```\n\nIt runs a random input through `weights\u002Fmb1_120x120.onnx` and prints the mean\nand standard deviation per inference.\n\n## ONNX conversion behavior\n\n- `FaceBoxes_ONNX` converts `FaceBoxesProd.pth` to `FaceBoxesProd.onnx` if the\n  ONNX file is missing.\n- `TDDFA_ONNX` converts the selected `.pth` checkpoint and the BFM decoder if\n  the ONNX artifacts are missing.\n- Conversion is CPU-safe but mutates the checkout by writing generated `.onnx`\n  files.\n\n## Thread and provider notes\n\n- The repo sets `OMP_NUM_THREADS=4` in the ONNX demo paths.\n- `KMP_DUPLICATE_LIB_OK=True` is set by the demo scripts before ONNX imports.\n- CPU ONNX Runtime is the baseline. CUDA\u002Fother providers are outside the\n  required verification scope.\n- On macOS, the original README notes that `libomp` may be required.\n\n## Troubleshooting\n\nRead `references\u002Ftroubleshooting.md` for ONNX-specific issues and\n`..\u002F..\u002Freferences\u002Ftroubleshooting.md` for shared setup failures.\n",{"name":5,"license":14,"metadata":23,"description":7,"disable-model-invocation":25},{"disco-role":24},"operating",true,"main","ac3fe1afa80fb9a09775ecfb2b6cc3ba850a2db6","skills\u002Frepositories\u002Frepo-skills\u002F3ddfa-v2\u002Fsub-skills\u002Fonnx-and-benchmarking",27,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill","https:\u002F\u002Fgithub.com\u002FVectorSpaceLab","2026-09-03 10:25:21",{"source":8,"skillName":5,"commands":34,"prompt":44,"zipAvailable":25},[35,38,41],{"tool":36,"command":37},"npx","npx skills add VectorSpaceLab\u002FAREX-Skill --skill onnx-and-benchmarking",{"tool":39,"command":40},"bunx","bunx skills add VectorSpaceLab\u002FAREX-Skill --skill onnx-and-benchmarking",{"tool":42,"command":43},"pnpm","pnpm dlx skills add VectorSpaceLab\u002FAREX-Skill --skill onnx-and-benchmarking","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002F3ddfa-v2\u002Fsub-skills\u002Fonnx-and-benchmarking\u002FSKILL.md",[46,52,56,62],{"path":47,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":50,"sourceHtmlUrl":51},"references\u002Ftroubleshooting.md","REFERENCE","参考文档",1202,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002F3ddfa-v2\u002Fsub-skills\u002Fonnx-and-benchmarking\u002Freferences\u002Ftroubleshooting.md",{"path":53,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":54,"sourceHtmlUrl":55},"references\u002Fworkflows.md",1147,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002F3ddfa-v2\u002Fsub-skills\u002Fonnx-and-benchmarking\u002Freferences\u002Fworkflows.md",{"path":57,"resourceType":58,"resourceTypeDesc":59,"sizeBytes":60,"sourceHtmlUrl":61},"scripts\u002Frun-latency.py","SCRIPT","脚本",910,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002F3ddfa-v2\u002Fsub-skills\u002Fonnx-and-benchmarking\u002Fscripts\u002Frun-latency.py",{"path":63,"resourceType":58,"resourceTypeDesc":59,"sizeBytes":64,"sourceHtmlUrl":65},"scripts\u002Frun-speed-cpu.py",920,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002F3ddfa-v2\u002Fsub-skills\u002Fonnx-and-benchmarking\u002Fscripts\u002Frun-speed-cpu.py",[67,76,83,90,97,104],{"skillNo":68,"name":69,"nameZh":6,"description":70,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":71,"licenseFileUrl":6,"sourceHtmlUrl":72,"categories":73,"keywords":74,"publishedAt":75},"01M799NYSH1R3RVAHSM2D88Y84","passnet-triton-opt","Make a single PassNet Triton kernel fast AND numerically faithful: performance model (when a replacement can win at all), block\u002Fgrid\u002Fwarp tuning, autotune policy, launch overhead, and per-op numeric recipes to pass the dtype baseline tolerances. Use when a pass matches and is correct but speedup below expected, or when correctness fails by small numeric margins.","Apache-2.0","https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Ftask-oriented\u002FPassNet\u002Fskills\u002Fpassnet-triton-opt\u002FSKILL.md",[],[],"2026-11-15 14:19:17",{"skillNo":77,"name":78,"nameZh":6,"description":79,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":71,"licenseFileUrl":6,"sourceHtmlUrl":80,"categories":81,"keywords":82,"publishedAt":75},"01M799NS5APEC0N1FZ7XZVW6BN","passnet-solve","END-TO-END playbook for solving one PassNet sample: analyze the computation graphs, decide the optimization strategy (what to fuse, what to replace, what to leave alone), drive the iteration loop, and maximize the sample score. This is the ENTRY skill — invoke it first for any PassNet optimization task; it tells you when to use passnet-pattern-fusion, passnet-triton-opt and passnet-feedback.","https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Ftask-oriented\u002FPassNet\u002Fskills\u002Fpassnet-solve\u002FSKILL.md",[],[],{"skillNo":84,"name":85,"nameZh":6,"description":86,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":71,"licenseFileUrl":6,"sourceHtmlUrl":87,"categories":88,"keywords":89,"publishedAt":75},"01M799NNY9BKSM21V5R7QVTBJX","passnet-skill","PassNet GPU kernel optimization via compiler passes. Design and implement Triton-based optimization passes, create pass files under .\u002Fpass_dir\u002F, self-evaluate with pass_evaluator, and iterate to maximize GPU speedup.","https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Ftask-oriented\u002FPassNet\u002Fskills\u002Fpassnet-skill\u002FSKILL.md",[],[],{"skillNo":91,"name":92,"nameZh":6,"description":93,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":71,"licenseFileUrl":6,"sourceHtmlUrl":94,"categories":95,"keywords":96,"publishedAt":75},"01M799NHH6NP7Z1RK77KGXAN6S","passnet-pattern-fusion","Author PassNet pass files: write patterns that actually MATCH the FX graph, pick fusion regions (which ops to absorb into one Triton kernel), and structure multi-pass submissions with the mandatory shared-dispatch architecture. Use when creating or fixing pattern()\u002Freplacement_args()\u002Freplacement_func() files, when a pass \"failed to match\", or when deciding how to fuse multiple kernels\u002Fops.","https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Ftask-oriented\u002FPassNet\u002Fskills\u002Fpassnet-pattern-fusion\u002FSKILL.md",[],[],{"skillNo":98,"name":99,"nameZh":6,"description":100,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":71,"licenseFileUrl":6,"sourceHtmlUrl":101,"categories":102,"keywords":103,"publishedAt":75},"01M799ND2J4WNC2PGAZ0A06NDF","passnet-orchestrate","Round\u002Fbatch planning and decision support for PassNet graph optimization. Use for multi-sample triage, worker allocation, eval-budget policy, and post-eval keep\u002Frevert decisions. For solving one concrete sample end to end, use passnet-solve as the entry skill; consult this skill only when you need broader planning or a specific decision gate.","https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Ftask-oriented\u002FPassNet\u002Fskills\u002Fpassnet-orchestrate\u002FSKILL.md",[],[],{"skillNo":105,"name":106,"nameZh":6,"description":107,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":71,"licenseFileUrl":6,"sourceHtmlUrl":108,"categories":109,"keywords":110,"publishedAt":75},"01M799N8SBTMY02ZDG2AZF5ZWE","passnet-feedback","Fast, deterministic feedback for PassNet work: pre-flight pattern\u002Fmatch verification WITHOUT burning a GPU evaluation, per-node bottleneck analysis of a sample's graphs, and eval-log parsing into per-variant status + estimated score + failure classification. Use BEFORE every GPU evaluation (check_pattern), at the START of a sample (analyze_graph), and AFTER every evaluation (parse_eval_log).","https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Ftask-oriented\u002FPassNet\u002Fskills\u002Fpassnet-feedback\u002FSKILL.md",[],[]]