[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M796K66NJSDEQWXYV3H5C3WF":3,"skill-same:01M796K66NJSDEQWXYV3H5C3WF":74},{"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},"01M796K66NJSDEQWXYV3H5C3WF","vision-data-models",null,"Operate TLLib vision datasets, ImageList formats, transforms, model factories, re-id metrics, and shared utilities.","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\u002Ftransfer-learning-library\u002Fsub-skills\u002Fvision-data-models\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: vision-data-models\ndescription: \"Operate TLLib vision datasets, ImageList formats, transforms,\n  model factories, re-id metrics, and shared utilities.\"\ndisable-model-invocation: true\nmetadata:\n  disco-role: operating\nlicense: MIT\n---\n\n# Vision Data and Models\n\nUse this sub-skill when a task needs TLLib's vision data\u002Fmodel layer rather than an algorithm-specific training recipe. It covers dataset classes, local image-list formats, classification\u002Fsegmentation\u002Fkeypoint\u002Fre-id\u002Fobject-detection model surfaces, transforms, metrics, data iterators, and logging helpers.\n\n## Route here for\n\n- Building or validating a local `ImageList` dataset for domain adaptation, domain generalization, task adaptation, or semi-supervised learning workflows.\n- Choosing a TLLib dataset wrapper such as Office31, OfficeHome, DomainNet, VisDA2017, PACS, digit datasets, natural object recognition datasets, segmentation datasets, keypoint datasets, regression datasets, or re-identification datasets.\n- Instantiating no-download model factories such as `resnet18`\u002F`resnet50`, `lenet`, `dtn`, `deeplabv2_resnet101`, `pose_resnet101`, and re-id ResNet\u002FReIdentifier components.\n- Using TLLib transforms, meters, `ForeverDataIterator`, `CombineDataset`, classification\u002Fsegmentation accuracy helpers, confusion matrices, keypoint metrics, or re-id CMC\u002FmAP utilities.\n- Diagnosing data-list, class-index, image-size\u002Fchannel, pretrained-weight, Torch\u002FTorchVision, object-detection, or re-id dependency failures.\n\n## Do not use this sub-skill for\n\n- DANN\u002FCDAN\u002FMK-MMD\u002FMDD\u002FCORAL\u002FOSBP\u002FMCC or other domain-adaptation losses and workflows; route to `domain-adaptation` after data\u002Fmodel setup.\n- MixStyle\u002FStochNorm\u002FIBN\u002FL2-SP\u002FDELTA\u002FBSS\u002FGroupDRO\u002Fdomain-generalization\u002Ffine-tuning workflows; route to `task-generalization` after data\u002Fmodel setup.\n- Pseudo-label, Mean Teacher, UDA, FixMatch, FlexMatch, or other self-training losses; route to `self-training` after data\u002Fmodel setup.\n- Transferability ranking metrics such as H-score, LEEP, NCE, LogME, or TransRate; route to `model-selection`.\n- CycleGAN\u002FFDA\u002FSPGAN\u002FCyCADA image translation internals; route to `translation`.\n\n## Operating sequence\n\n1. Identify the task family and whether the user has local data, external data to obtain under its own license, or only a synthetic\u002Fsmoke fixture.\n2. For local classification-style data, validate the image-list file with `scripts\u002Fvalidate_imagelist.py` before wiring it into a training command.\n3. Choose data transforms and model factories from `references\u002Fvision-data-and-models.md`; avoid pretrained downloads unless the user explicitly allows network\u002Fcache use.\n4. For re-identification, use `references\u002Freid-and-metrics.md` for dataset tuple conventions, model\u002Floss surfaces, and CMC\u002FmAP behavior.\n5. For failures, start with `references\u002Ftroubleshooting.md` before changing dependencies or dataset layout.\n6. For an import\u002Fmodel\u002Fdata smoke check in an installed environment, run `scripts\u002Ftllib_vision_smoke.py`.\n\n## Bundled references\n\n- `references\u002Fdatasets-and-formats.md` - dataset classes, no-download policy, ImageList\u002FMultipleDomainsDataset\u002FSegmentationList\u002Fre-id\u002Fkeypoint\u002Fregression format contracts.\n- `references\u002Fvision-data-and-models.md` - transforms, model factories, task-specific model surfaces, utilities, meters, loggers, metrics.\n- `references\u002Freid-and-metrics.md` - person re-id datasets, samplers, models, losses, CMC\u002FmAP\u002Fre-ranking, validation conventions.\n- `references\u002Ftroubleshooting.md` - broken links, licenses, path\u002Flabel mistakes, TorchVision compatibility, pretrained downloads, image shape\u002Fchannel issues, optional dependency problems.\n\n## Bundled scripts\n\n- `scripts\u002Fvalidate_imagelist.py`: safe local `ImageList` format validator. It does not download data and can create a TLLib `ImageList` object from an installed `tllib` package.\n- `scripts\u002Ftllib_vision_smoke.py`: no-network CPU smoke for installed `tllib` vision datasets, transforms, models, metrics, and utility imports. Object-detection imports are reported as optional if Detectron2 is unavailable.\n","# Vision Data and Models\n\nUse this sub-skill when a task needs TLLib's vision data\u002Fmodel layer rather than an algorithm-specific training recipe. It covers dataset classes, local image-list formats, classification\u002Fsegmentation\u002Fkeypoint\u002Fre-id\u002Fobject-detection model surfaces, transforms, metrics, data iterators, and logging helpers.\n\n## Route here for\n\n- Building or validating a local `ImageList` dataset for domain adaptation, domain generalization, task adaptation, or semi-supervised learning workflows.\n- Choosing a TLLib dataset wrapper such as Office31, OfficeHome, DomainNet, VisDA2017, PACS, digit datasets, natural object recognition datasets, segmentation datasets, keypoint datasets, regression datasets, or re-identification datasets.\n- Instantiating no-download model factories such as `resnet18`\u002F`resnet50`, `lenet`, `dtn`, `deeplabv2_resnet101`, `pose_resnet101`, and re-id ResNet\u002FReIdentifier components.\n- Using TLLib transforms, meters, `ForeverDataIterator`, `CombineDataset`, classification\u002Fsegmentation accuracy helpers, confusion matrices, keypoint metrics, or re-id CMC\u002FmAP utilities.\n- Diagnosing data-list, class-index, image-size\u002Fchannel, pretrained-weight, Torch\u002FTorchVision, object-detection, or re-id dependency failures.\n\n## Do not use this sub-skill for\n\n- DANN\u002FCDAN\u002FMK-MMD\u002FMDD\u002FCORAL\u002FOSBP\u002FMCC or other domain-adaptation losses and workflows; route to `domain-adaptation` after data\u002Fmodel setup.\n- MixStyle\u002FStochNorm\u002FIBN\u002FL2-SP\u002FDELTA\u002FBSS\u002FGroupDRO\u002Fdomain-generalization\u002Ffine-tuning workflows; route to `task-generalization` after data\u002Fmodel setup.\n- Pseudo-label, Mean Teacher, UDA, FixMatch, FlexMatch, or other self-training losses; route to `self-training` after data\u002Fmodel setup.\n- Transferability ranking metrics such as H-score, LEEP, NCE, LogME, or TransRate; route to `model-selection`.\n- CycleGAN\u002FFDA\u002FSPGAN\u002FCyCADA image translation internals; route to `translation`.\n\n## Operating sequence\n\n1. Identify the task family and whether the user has local data, external data to obtain under its own license, or only a synthetic\u002Fsmoke fixture.\n2. For local classification-style data, validate the image-list file with `scripts\u002Fvalidate_imagelist.py` before wiring it into a training command.\n3. Choose data transforms and model factories from `references\u002Fvision-data-and-models.md`; avoid pretrained downloads unless the user explicitly allows network\u002Fcache use.\n4. For re-identification, use `references\u002Freid-and-metrics.md` for dataset tuple conventions, model\u002Floss surfaces, and CMC\u002FmAP behavior.\n5. For failures, start with `references\u002Ftroubleshooting.md` before changing dependencies or dataset layout.\n6. For an import\u002Fmodel\u002Fdata smoke check in an installed environment, run `scripts\u002Ftllib_vision_smoke.py`.\n\n## Bundled references\n\n- `references\u002Fdatasets-and-formats.md` - dataset classes, no-download policy, ImageList\u002FMultipleDomainsDataset\u002FSegmentationList\u002Fre-id\u002Fkeypoint\u002Fregression format contracts.\n- `references\u002Fvision-data-and-models.md` - transforms, model factories, task-specific model surfaces, utilities, meters, loggers, metrics.\n- `references\u002Freid-and-metrics.md` - person re-id datasets, samplers, models, losses, CMC\u002FmAP\u002Fre-ranking, validation conventions.\n- `references\u002Ftroubleshooting.md` - broken links, licenses, path\u002Flabel mistakes, TorchVision compatibility, pretrained downloads, image shape\u002Fchannel issues, optional dependency problems.\n\n## Bundled scripts\n\n- `scripts\u002Fvalidate_imagelist.py`: safe local `ImageList` format validator. It does not download data and can create a TLLib `ImageList` object from an installed `tllib` package.\n- `scripts\u002Ftllib_vision_smoke.py`: no-network CPU smoke for installed `tllib` vision datasets, transforms, models, metrics, and utility imports. Object-detection imports are reported as optional if Detectron2 is unavailable.\n",{"name":5,"license":14,"metadata":23,"description":7,"disable-model-invocation":25},{"disco-role":24},"operating",true,"main","ac3fe1afa80fb9a09775ecfb2b6cc3ba850a2db6","skills\u002Frepositories\u002Frepo-skills\u002Ftransfer-learning-library\u002Fsub-skills\u002Fvision-data-models",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 vision-data-models",{"tool":39,"command":40},"bunx","bunx skills add VectorSpaceLab\u002FAREX-Skill --skill vision-data-models",{"tool":42,"command":43},"pnpm","pnpm dlx skills add VectorSpaceLab\u002FAREX-Skill --skill vision-data-models","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Ftransfer-learning-library\u002Fsub-skills\u002Fvision-data-models\u002FSKILL.md",[46,52,56,60,64,70],{"path":47,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":50,"sourceHtmlUrl":51},"references\u002Fdatasets-and-formats.md","REFERENCE","参考文档",9586,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Ftransfer-learning-library\u002Fsub-skills\u002Fvision-data-models\u002Freferences\u002Fdatasets-and-formats.md",{"path":53,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":54,"sourceHtmlUrl":55},"references\u002Freid-and-metrics.md",7018,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Ftransfer-learning-library\u002Fsub-skills\u002Fvision-data-models\u002Freferences\u002Freid-and-metrics.md",{"path":57,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":58,"sourceHtmlUrl":59},"references\u002Ftroubleshooting.md",9577,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Ftransfer-learning-library\u002Fsub-skills\u002Fvision-data-models\u002Freferences\u002Ftroubleshooting.md",{"path":61,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":62,"sourceHtmlUrl":63},"references\u002Fvision-data-and-models.md",10553,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Ftransfer-learning-library\u002Fsub-skills\u002Fvision-data-models\u002Freferences\u002Fvision-data-and-models.md",{"path":65,"resourceType":66,"resourceTypeDesc":67,"sizeBytes":68,"sourceHtmlUrl":69},"scripts\u002Ftllib_vision_smoke.py","SCRIPT","脚本",9897,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Ftransfer-learning-library\u002Fsub-skills\u002Fvision-data-models\u002Fscripts\u002Ftllib_vision_smoke.py",{"path":71,"resourceType":66,"resourceTypeDesc":67,"sizeBytes":72,"sourceHtmlUrl":73},"scripts\u002Fvalidate_imagelist.py",8626,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Ftransfer-learning-library\u002Fsub-skills\u002Fvision-data-models\u002Fscripts\u002Fvalidate_imagelist.py",[75,83,90,97,104,111],{"skillNo":76,"name":77,"nameZh":6,"description":78,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":79,"licenseFileUrl":6,"sourceHtmlUrl":80,"categories":81,"keywords":82,"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.","Apache-2.0","https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Ftask-oriented\u002FPassNet\u002Fskills\u002Fpassnet-triton-opt\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":79,"licenseFileUrl":6,"sourceHtmlUrl":87,"categories":88,"keywords":89,"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":91,"name":92,"nameZh":6,"description":93,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":79,"licenseFileUrl":6,"sourceHtmlUrl":94,"categories":95,"keywords":96,"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":98,"name":99,"nameZh":6,"description":100,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":79,"licenseFileUrl":6,"sourceHtmlUrl":101,"categories":102,"keywords":103,"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":105,"name":106,"nameZh":6,"description":107,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":79,"licenseFileUrl":6,"sourceHtmlUrl":108,"categories":109,"keywords":110,"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":112,"name":113,"nameZh":6,"description":114,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":79,"licenseFileUrl":6,"sourceHtmlUrl":115,"categories":116,"keywords":117,"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",[],[]]