[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M78YBEREMNTS5YCMSYB5F7RN":3,"skill-same:01M78YBEREMNTS5YCMSYB5F7RN":70},{"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},"01M78YBEREMNTS5YCMSYB5F7RN","training-evaluation",null,"Build safe dataset-aware training and evaluation commands for the verified Gluon and PyTorch classification 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\u002Fimgclsmob\u002Fsub-skills\u002Ftraining-evaluation\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: training-evaluation\ndescription: \"Build safe dataset-aware training and evaluation commands for the\n  verified Gluon and PyTorch classification workflows.\"\ndisable-model-invocation: true\nmetadata:\n  disco-role: operating\nlicense: MIT\n---\n\n# Training and evaluation\n\nUse this sub-skill when the request concerns ImageNet1K, CIFAR10, CIFAR100,\nSVHN, or CUB200_2011 data layout, `train_gl.py`\u002F`eval_gl.py`,\n`train_pt.py`\u002F`eval_pt.py`, metrics, batch\u002Fdevice sizing, or checkpoint\nresume behavior. This sub-skill documents classification workflows only; do\nnot turn it into a full training run.\n\n## Route first\n\n- Use [model-inference](..\u002Fmodel-inference\u002FSKILL.md) to choose and construct a\n  model, select `resnet18`-style names, or diagnose model\u002Fcheckpoint shape and\n  device loading.\n- Use [framework-compatibility](..\u002Fframework-compatibility\u002FSKILL.md) for\n  TensorFlow, Keras, Chainer, legacy Tensorpack, or other optional backend\n  claims. CPU Gluon and CPU PyTorch\u002Fpytorchcv are the verified core; do not\n  infer CUDA or optional-backend support from this sub-skill.\n\n## Safe operating procedure\n\n1. Select the exact dataset name and layout. `ImageNet1K` means folder layout;\n   `ImageNet1K_rec` is the Gluon-only MXNet record layout. Read\n   [datasets and layouts](references\u002Fdatasets-and-layouts.md).\n2. Run the no-network preflight before a real command:\n\n   ```bash\n   python scripts\u002Fcheck_dataset_layout.py --dataset ImageNet1K --data-dir \u002Fdata\u002Fimagenet --backend pytorch\n   ```\n\n   Replace `\u002Fdata\u002Fimagenet` with the user's local data root. The checker only\n   inspects filesystem structure; it never downloads data or imports a\n   framework. A nonzero result means stop and fix the root\u002Flayout rather than\n   launching a CLI.\n3. Construct the command from [the CLI reference](references\u002Fcli-reference.md).\n   Start with `--num-gpus=0`, `--num-data-workers 0`, and a conservative\n   `--batch-size` on CPU. Omit `--use-pretrained`, `--all`, and any other\n   network-dependent option for offline work. For CIFAR10, CIFAR100, and\n   SVHN, an `indeterminate_native_cache` result means only that a local cache\n   exists; the selected framework must still verify its cache with downloads\n   disabled.\n4. Run `--help` first. Help parsing does not construct a model, open a data\n   loader, download weights, or start training. Do not use help output as proof\n   that a dataset or checkpoint is valid.\n5. For evaluation, prefer `--data-subset val` for these classification\n   metainfo objects. The scripts use the held-out split for `val`; several\n   listed metainfo objects do not define separate `test_metric_names`.\n6. Record the selected dataset, root, framework, checkpoint type, device,\n   effective batch size, metric names, and validation result. Do not claim an\n   accuracy value without running the evaluation against a real local dataset.\n\n## Checkpoint and command boundaries\n\n- Gluon training separates model parameters (`--resume`, normally a\n  `.params` file) from optimizer state (`--resume-state`, normally a `.states`\n  file). PyTorch evaluation loads a model checkpoint through `--resume`,\n  normally `.pth`; PyTorch training uses `--resume-state` for a state object\n  containing `epoch`, `state_dict`, and `optimizer`.\n- `--start-epoch` is an explicit 1-based control in both training CLIs. Passing\n  a state file does not remove the need to choose the intended start epoch.\n- A local `--resume` is distinct from `--use-pretrained`: the latter can ask a\n  model provider to obtain weights. For no-network operation, use a local\n  checkpoint and omit `--use-pretrained`.\n- Use [troubleshooting](references\u002Ftroubleshooting.md) when the layout,\n  metric, checkpoint, or CPU\u002Fdevice assumptions fail.\n\n## Output contract\n\nReturn a command plan, not a long-running experiment: dataset\u002Flayout decision,\npreflight result, exact CLI flags, expected metric\u002Fcheckpoint behavior, and the\nnext bounded check. Link model construction and optional backend questions as\nspecified above.\n","# Training and evaluation\n\nUse this sub-skill when the request concerns ImageNet1K, CIFAR10, CIFAR100,\nSVHN, or CUB200_2011 data layout, `train_gl.py`\u002F`eval_gl.py`,\n`train_pt.py`\u002F`eval_pt.py`, metrics, batch\u002Fdevice sizing, or checkpoint\nresume behavior. This sub-skill documents classification workflows only; do\nnot turn it into a full training run.\n\n## Route first\n\n- Use [model-inference](..\u002Fmodel-inference\u002FSKILL.md) to choose and construct a\n  model, select `resnet18`-style names, or diagnose model\u002Fcheckpoint shape and\n  device loading.\n- Use [framework-compatibility](..\u002Fframework-compatibility\u002FSKILL.md) for\n  TensorFlow, Keras, Chainer, legacy Tensorpack, or other optional backend\n  claims. CPU Gluon and CPU PyTorch\u002Fpytorchcv are the verified core; do not\n  infer CUDA or optional-backend support from this sub-skill.\n\n## Safe operating procedure\n\n1. Select the exact dataset name and layout. `ImageNet1K` means folder layout;\n   `ImageNet1K_rec` is the Gluon-only MXNet record layout. Read\n   [datasets and layouts](references\u002Fdatasets-and-layouts.md).\n2. Run the no-network preflight before a real command:\n\n   ```bash\n   python scripts\u002Fcheck_dataset_layout.py --dataset ImageNet1K --data-dir \u002Fdata\u002Fimagenet --backend pytorch\n   ```\n\n   Replace `\u002Fdata\u002Fimagenet` with the user's local data root. The checker only\n   inspects filesystem structure; it never downloads data or imports a\n   framework. A nonzero result means stop and fix the root\u002Flayout rather than\n   launching a CLI.\n3. Construct the command from [the CLI reference](references\u002Fcli-reference.md).\n   Start with `--num-gpus=0`, `--num-data-workers 0`, and a conservative\n   `--batch-size` on CPU. Omit `--use-pretrained`, `--all`, and any other\n   network-dependent option for offline work. For CIFAR10, CIFAR100, and\n   SVHN, an `indeterminate_native_cache` result means only that a local cache\n   exists; the selected framework must still verify its cache with downloads\n   disabled.\n4. Run `--help` first. Help parsing does not construct a model, open a data\n   loader, download weights, or start training. Do not use help output as proof\n   that a dataset or checkpoint is valid.\n5. For evaluation, prefer `--data-subset val` for these classification\n   metainfo objects. The scripts use the held-out split for `val`; several\n   listed metainfo objects do not define separate `test_metric_names`.\n6. Record the selected dataset, root, framework, checkpoint type, device,\n   effective batch size, metric names, and validation result. Do not claim an\n   accuracy value without running the evaluation against a real local dataset.\n\n## Checkpoint and command boundaries\n\n- Gluon training separates model parameters (`--resume`, normally a\n  `.params` file) from optimizer state (`--resume-state`, normally a `.states`\n  file). PyTorch evaluation loads a model checkpoint through `--resume`,\n  normally `.pth`; PyTorch training uses `--resume-state` for a state object\n  containing `epoch`, `state_dict`, and `optimizer`.\n- `--start-epoch` is an explicit 1-based control in both training CLIs. Passing\n  a state file does not remove the need to choose the intended start epoch.\n- A local `--resume` is distinct from `--use-pretrained`: the latter can ask a\n  model provider to obtain weights. For no-network operation, use a local\n  checkpoint and omit `--use-pretrained`.\n- Use [troubleshooting](references\u002Ftroubleshooting.md) when the layout,\n  metric, checkpoint, or CPU\u002Fdevice assumptions fail.\n\n## Output contract\n\nReturn a command plan, not a long-running experiment: dataset\u002Flayout decision,\npreflight result, exact CLI flags, expected metric\u002Fcheckpoint behavior, and the\nnext bounded check. Link model construction and optional backend questions as\nspecified above.\n",{"name":5,"license":14,"metadata":23,"description":7,"disable-model-invocation":25},{"disco-role":24},"operating",true,"main","ac3fe1afa80fb9a09775ecfb2b6cc3ba850a2db6","skills\u002Frepositories\u002Frepo-skills\u002Fimgclsmob\u002Fsub-skills\u002Ftraining-evaluation",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 training-evaluation",{"tool":39,"command":40},"bunx","bunx skills add VectorSpaceLab\u002FAREX-Skill --skill training-evaluation",{"tool":42,"command":43},"pnpm","pnpm dlx skills add VectorSpaceLab\u002FAREX-Skill --skill training-evaluation","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fimgclsmob\u002Fsub-skills\u002Ftraining-evaluation\u002FSKILL.md",[46,52,56,60,66],{"path":47,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":50,"sourceHtmlUrl":51},"references\u002Fcli-reference.md","REFERENCE","参考文档",12339,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fimgclsmob\u002Fsub-skills\u002Ftraining-evaluation\u002Freferences\u002Fcli-reference.md",{"path":53,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":54,"sourceHtmlUrl":55},"references\u002Fdatasets-and-layouts.md",5885,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fimgclsmob\u002Fsub-skills\u002Ftraining-evaluation\u002Freferences\u002Fdatasets-and-layouts.md",{"path":57,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":58,"sourceHtmlUrl":59},"references\u002Ftroubleshooting.md",10533,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fimgclsmob\u002Fsub-skills\u002Ftraining-evaluation\u002Freferences\u002Ftroubleshooting.md",{"path":61,"resourceType":62,"resourceTypeDesc":63,"sizeBytes":64,"sourceHtmlUrl":65},"scripts\u002Fbuild_command.py","SCRIPT","脚本",4182,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fimgclsmob\u002Fsub-skills\u002Ftraining-evaluation\u002Fscripts\u002Fbuild_command.py",{"path":67,"resourceType":62,"resourceTypeDesc":63,"sizeBytes":68,"sourceHtmlUrl":69},"scripts\u002Fcheck_dataset_layout.py",11759,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fimgclsmob\u002Fsub-skills\u002Ftraining-evaluation\u002Fscripts\u002Fcheck_dataset_layout.py",[71,79,86,93,100,107],{"skillNo":72,"name":73,"nameZh":6,"description":74,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":75,"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.","Apache-2.0","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":75,"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":75,"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":75,"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":75,"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":75,"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",[],[]]