[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M790MYMW72MS4016QQE55ZPJ":3,"skill-same:01M790MYMW72MS4016QQE55ZPJ":71},{"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":27,"commitSha":28,"skillDirectory":29,"forksCount":30,"repositoryHtmlUrl":31,"repositoryHomepageUrl":6,"ownerHtmlUrl":32,"ownerBlogUrl":6,"ownerDescription":6,"updatedAt":33,"install":34,"resources":46},"01M790MYMW72MS4016QQE55ZPJ","mmpretrain",null,"Operate MMPreTrain model-zoo inference, config-driven training and evaluation, dataset\u002Fcustomization, analysis utilities, checkpoint conversion, and serving workflows.","VectorSpaceLab\u002FAREX-Skill","VectorSpaceLab","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F180637102?v=4","01M7538C5WPSJGWF7KHR4EFZ9F","01M7538C6R2GFP53NPXTCZNBRT",351,"Apache-2.0","https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fmmpretrain\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: mmpretrain\ndescription: \"Operate MMPreTrain model-zoo inference, config-driven training and\n  evaluation, dataset\u002Fcustomization, analysis utilities, checkpoint conversion,\n  and serving workflows.\"\ndisable-model-invocation: true\nmetadata:\n  disco-role: operating\nlicense: Apache 2.0\n---\n\n# MMPreTrain repo skill\n\nUse this skill when a task involves MMPreTrain \u002F OpenMMLab Model Pretraining Toolbox as a package or framework: model-zoo discovery, inference, feature extraction, config-driven training\u002Ftesting, dataset preparation, registry customization, analysis tools, checkpoint conversion, or serving packaging.\n\n## First checks\n\n- Read `references\u002Fpackage-overview.md` for the package surfaces and optional capability map.\n- Read `references\u002Finstallation-and-troubleshooting.md` when installation, import, ModelHub, optional dependencies, or backend readiness are uncertain.\n- Run `scripts\u002Fcheck_mmpretrain_env.py` for a no-download package smoke check before relying on model-zoo APIs or backend-specific behavior.\n- Read `references\u002Frepo-provenance.md` before deciding whether this skill is stale for a different checkout or package version.\n\n## Route by task\n\n| User task | Read |\n| --- | --- |\n| List available models, choose a checkpoint policy, use `get_model`, run image classification, retrieval, captioning, VQA, visual grounding, NLVR, or extract features | `sub-skills\u002Fmodel-zoo-inference\u002FSKILL.md` |\n| Inspect or modify configs, plan `mim train mmpretrain` \u002F `mim test mmpretrain`, choose CPU\u002FGPU\u002Fdistributed\u002FSlurm launch, use resume\u002FAMP\u002Fauto-scale-lr\u002FTTA\u002Foutput dumps, or plan K-fold | `sub-skills\u002Ftraining-and-evaluation\u002FSKILL.md` |\n| Prepare `CustomDataset`, ImageNet, or OpenMMLab 2.0 annotations; inspect dataloader fields; build transform pipelines; register custom datasets\u002Fmodels\u002Fmetrics\u002Fhooks\u002Foptimizers | `sub-skills\u002Fdatasets-and-customization\u002FSKILL.md` |\n| Analyze JSON logs\u002Fresults, compute metrics\u002Fconfusion matrices\u002FFLOPs, visualize CAM\u002Ft-SNE\u002Fschedulers\u002Fdatasets, publish\u002Fconvert checkpoints, reparameterize, or package TorchServe artifacts | `sub-skills\u002Ftools-analysis-and-deployment\u002FSKILL.md` |\n\n## Install baseline\n\nTypical package install:\n\n```bash\npip install -U openmim\nmim install \"mmpretrain>=1.0.0rc8\"\n```\n\nDevelopment\u002Fpackage-tool install:\n\n```bash\npip install -U openmim\nmim install -e .\n```\n\nMulti-modal workflows need the optional dependency group:\n\n```bash\nmim install \"mmpretrain[multimodal]>=1.0.0rc8\"\n```\n\nMMPreTrain expects Python 3.7+, PyTorch 1.8+, `mmcv>=2.0.0,\u003C2.4.0`, and `mmengine>=0.8.3,\u003C1.0.0`. Select a CUDA-capable PyTorch\u002FMMCV build only when the task requires GPU or distributed behavior.\n\n## Safe operating defaults\n\n1. For offline or architecture-only model work, use `get_model(..., pretrained=False)` or inferencer `pretrained=False`; inferencer classes otherwise try default weights.\n2. For CPU-only train\u002Ftest planning, use MIM package commands with `--gpus 0` and\u002For `CUDA_VISIBLE_DEVICES=-1`.\n3. Inspect a merged config and dataset fields before long runs; route config work to `training-and-evaluation` and dataset schema work to `datasets-and-customization`.\n4. Treat checkpoint downloads, dataset downloads, TorchServe startup, Slurm jobs, and GPU training as resource-requiring actions. Confirm resources before launching them.\n5. Use bundled helpers and references here; do not depend on the original repository checkout being present.\n\n## Common failure routing\n\n- `File ... .mim\u002Fmodel-index.yml does not exist`, incompatible `mmcv` \u002F `mmengine`, NumPy ABI warnings, or CUDA backend mismatch: `references\u002Finstallation-and-troubleshooting.md`.\n- Invalid model name, checkpoint download\u002Fcache error, device mismatch, or missing multimodal extras during inference: `sub-skills\u002Fmodel-zoo-inference\u002Freferences\u002Ftroubleshooting.md`.\n- Bad `_base_` inheritance, `--cfg-options` quoting, `load_from` versus `resume`, AMP\u002FTTA, or distributed launch hangs: `sub-skills\u002Ftraining-and-evaluation\u002Freferences\u002Ftroubleshooting.md`.\n- Bad annotation paths\u002Flabels, missing `LoadImageFromFile`, unreadable images, or registry `KeyError`: `sub-skills\u002Fdatasets-and-customization\u002Freferences\u002Ftroubleshooting.md`.\n- Plotting\u002Fheadless errors, optional Grad-CAM\u002Ft-SNE deps, checkpoint format mismatch, EMA publishing, or TorchServe service errors: `sub-skills\u002Ftools-analysis-and-deployment\u002Freferences\u002Ftroubleshooting.md`.\n","# MMPreTrain repo skill\n\nUse this skill when a task involves MMPreTrain \u002F OpenMMLab Model Pretraining Toolbox as a package or framework: model-zoo discovery, inference, feature extraction, config-driven training\u002Ftesting, dataset preparation, registry customization, analysis tools, checkpoint conversion, or serving packaging.\n\n## First checks\n\n- Read `references\u002Fpackage-overview.md` for the package surfaces and optional capability map.\n- Read `references\u002Finstallation-and-troubleshooting.md` when installation, import, ModelHub, optional dependencies, or backend readiness are uncertain.\n- Run `scripts\u002Fcheck_mmpretrain_env.py` for a no-download package smoke check before relying on model-zoo APIs or backend-specific behavior.\n- Read `references\u002Frepo-provenance.md` before deciding whether this skill is stale for a different checkout or package version.\n\n## Route by task\n\n| User task | Read |\n| --- | --- |\n| List available models, choose a checkpoint policy, use `get_model`, run image classification, retrieval, captioning, VQA, visual grounding, NLVR, or extract features | `sub-skills\u002Fmodel-zoo-inference\u002FSKILL.md` |\n| Inspect or modify configs, plan `mim train mmpretrain` \u002F `mim test mmpretrain`, choose CPU\u002FGPU\u002Fdistributed\u002FSlurm launch, use resume\u002FAMP\u002Fauto-scale-lr\u002FTTA\u002Foutput dumps, or plan K-fold | `sub-skills\u002Ftraining-and-evaluation\u002FSKILL.md` |\n| Prepare `CustomDataset`, ImageNet, or OpenMMLab 2.0 annotations; inspect dataloader fields; build transform pipelines; register custom datasets\u002Fmodels\u002Fmetrics\u002Fhooks\u002Foptimizers | `sub-skills\u002Fdatasets-and-customization\u002FSKILL.md` |\n| Analyze JSON logs\u002Fresults, compute metrics\u002Fconfusion matrices\u002FFLOPs, visualize CAM\u002Ft-SNE\u002Fschedulers\u002Fdatasets, publish\u002Fconvert checkpoints, reparameterize, or package TorchServe artifacts | `sub-skills\u002Ftools-analysis-and-deployment\u002FSKILL.md` |\n\n## Install baseline\n\nTypical package install:\n\n```bash\npip install -U openmim\nmim install \"mmpretrain>=1.0.0rc8\"\n```\n\nDevelopment\u002Fpackage-tool install:\n\n```bash\npip install -U openmim\nmim install -e .\n```\n\nMulti-modal workflows need the optional dependency group:\n\n```bash\nmim install \"mmpretrain[multimodal]>=1.0.0rc8\"\n```\n\nMMPreTrain expects Python 3.7+, PyTorch 1.8+, `mmcv>=2.0.0,\u003C2.4.0`, and `mmengine>=0.8.3,\u003C1.0.0`. Select a CUDA-capable PyTorch\u002FMMCV build only when the task requires GPU or distributed behavior.\n\n## Safe operating defaults\n\n1. For offline or architecture-only model work, use `get_model(..., pretrained=False)` or inferencer `pretrained=False`; inferencer classes otherwise try default weights.\n2. For CPU-only train\u002Ftest planning, use MIM package commands with `--gpus 0` and\u002For `CUDA_VISIBLE_DEVICES=-1`.\n3. Inspect a merged config and dataset fields before long runs; route config work to `training-and-evaluation` and dataset schema work to `datasets-and-customization`.\n4. Treat checkpoint downloads, dataset downloads, TorchServe startup, Slurm jobs, and GPU training as resource-requiring actions. Confirm resources before launching them.\n5. Use bundled helpers and references here; do not depend on the original repository checkout being present.\n\n## Common failure routing\n\n- `File ... .mim\u002Fmodel-index.yml does not exist`, incompatible `mmcv` \u002F `mmengine`, NumPy ABI warnings, or CUDA backend mismatch: `references\u002Finstallation-and-troubleshooting.md`.\n- Invalid model name, checkpoint download\u002Fcache error, device mismatch, or missing multimodal extras during inference: `sub-skills\u002Fmodel-zoo-inference\u002Freferences\u002Ftroubleshooting.md`.\n- Bad `_base_` inheritance, `--cfg-options` quoting, `load_from` versus `resume`, AMP\u002FTTA, or distributed launch hangs: `sub-skills\u002Ftraining-and-evaluation\u002Freferences\u002Ftroubleshooting.md`.\n- Bad annotation paths\u002Flabels, missing `LoadImageFromFile`, unreadable images, or registry `KeyError`: `sub-skills\u002Fdatasets-and-customization\u002Freferences\u002Ftroubleshooting.md`.\n- Plotting\u002Fheadless errors, optional Grad-CAM\u002Ft-SNE deps, checkpoint format mismatch, EMA publishing, or TorchServe service errors: `sub-skills\u002Ftools-analysis-and-deployment\u002Freferences\u002Ftroubleshooting.md`.\n",{"name":5,"license":23,"metadata":24,"description":7,"disable-model-invocation":26},"Apache 2.0",{"disco-role":25},"operating",true,"main","ac3fe1afa80fb9a09775ecfb2b6cc3ba850a2db6","skills\u002Frepositories\u002Frepo-skills\u002Fmmpretrain",27,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill","https:\u002F\u002Fgithub.com\u002FVectorSpaceLab","2026-09-03 10:25:21",{"source":8,"skillName":5,"commands":35,"prompt":45,"zipAvailable":26},[36,39,42],{"tool":37,"command":38},"npx","npx skills add VectorSpaceLab\u002FAREX-Skill --skill mmpretrain",{"tool":40,"command":41},"bunx","bunx skills add VectorSpaceLab\u002FAREX-Skill --skill mmpretrain",{"tool":43,"command":44},"pnpm","pnpm dlx skills add VectorSpaceLab\u002FAREX-Skill --skill mmpretrain","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fmmpretrain\u002FSKILL.md",[47,53,57,61,65],{"path":48,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":51,"sourceHtmlUrl":52},"references\u002Finstallation-and-troubleshooting.md","REFERENCE","参考文档",4441,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fmmpretrain\u002Freferences\u002Finstallation-and-troubleshooting.md",{"path":54,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":55,"sourceHtmlUrl":56},"references\u002Fpackage-overview.md",3149,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fmmpretrain\u002Freferences\u002Fpackage-overview.md",{"path":58,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":59,"sourceHtmlUrl":60},"references\u002Frepo-provenance.md",1822,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fmmpretrain\u002Freferences\u002Frepo-provenance.md",{"path":62,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":63,"sourceHtmlUrl":64},"references\u002Frepo-routing-metadata.json",510,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fmmpretrain\u002Freferences\u002Frepo-routing-metadata.json",{"path":66,"resourceType":67,"resourceTypeDesc":68,"sizeBytes":69,"sourceHtmlUrl":70},"scripts\u002Fcheck_mmpretrain_env.py","SCRIPT","脚本",5213,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fmmpretrain\u002Fscripts\u002Fcheck_mmpretrain_env.py",[72,79,86,93,100,107],{"skillNo":73,"name":74,"nameZh":6,"description":75,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":76,"categories":77,"keywords":78,"publishedAt":18},"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.","https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Ftask-oriented\u002FPassNet\u002Fskills\u002Fpassnet-triton-opt\u002FSKILL.md",[],[],{"skillNo":80,"name":81,"nameZh":6,"description":82,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":83,"categories":84,"keywords":85,"publishedAt":18},"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":87,"name":88,"nameZh":6,"description":89,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":90,"categories":91,"keywords":92,"publishedAt":18},"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":94,"name":95,"nameZh":6,"description":96,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":97,"categories":98,"keywords":99,"publishedAt":18},"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":101,"name":102,"nameZh":6,"description":103,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":104,"categories":105,"keywords":106,"publishedAt":18},"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":108,"name":109,"nameZh":6,"description":110,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":111,"categories":112,"keywords":113,"publishedAt":18},"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",[],[]]