[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M796D9GFWVECSQR46CNHTH9M":3,"skill-same:01M796D9GFWVECSQR46CNHTH9M":52},{"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},"01M796D9GFWVECSQR46CNHTH9M","datamodules-and-tasks",null,"Use for TorchGeo Lightning DataModules, CLI\u002Fconfig runs, task classes, losses, metrics, prediction\u002Fplot hooks, and training workflow wiring.","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\u002Ftorchgeo\u002Fsub-skills\u002Fdatamodules-and-tasks\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: datamodules-and-tasks\ndescription: \"Use for TorchGeo Lightning DataModules, CLI\u002Fconfig runs, task\n  classes, losses, metrics, prediction\u002Fplot hooks, and training workflow\n  wiring.\"\nmetadata:\n  disco-role: operating\ndisable-model-invocation: true\nlicense: MIT\n---\n\n# TorchGeo datamodules and tasks\n\nUse this sub-skill when wiring TorchGeo into Lightning training, validation, testing, prediction, CLI configs, or task modules.\n\n## DataModule operating facts\n\n- `BaseDataModule` wraps a dataset class, `batch_size`, `num_workers`, split datasets, split dataloaders, and Kornia augmentation hooks.\n- `prepare_data()` may download data when `download=True`; it must not create split state. `setup(stage)` creates datasets\u002Fsamplers and is called on every process.\n- `GeoDataModule` is for `GeoDataset` classes and sets `collate_fn = stack_samples`. It manages `RandomPatchSampler`, `GriddedPatchSampler`, and batch samplers for split-specific patch extraction.\n- `NonGeoDataModule` is for integer-indexed datasets and ordinary PyTorch dataloader shuffle\u002Fsplit semantics.\n- `on_after_batch_transfer()` applies the split-specific augmentation after the batch reaches the device. This is where Kornia GPU augmentations belong.\n\n## Task classes\n\nTorchGeo task modules are Lightning modules for common geospatial ML problems:\n\n- `Classification`: timm-backed image classification with `weights`, `in_channels`, task type, class\u002Flabel counts, losses, and optional frozen backbone.\n- `SemanticSegmentation`: SMP or TorchGeo FCN segmentation with `model`, `backbone`, `weights`, `in_channels`, multiclass\u002Fmultilabel\u002Fbinary settings, losses, and freeze options.\n- Other task modules cover change detection, detection, instance segmentation, regression, temporal\u002Fspatiotemporal tasks, MAE, MoCo, BYOL, SimCLR, and IO benchmark workflows.\n\nBefore changing a task, inspect its corresponding `tests\u002Ftasks\u002Ftest_\u003Ctask>.py` for constructor matrix, batch schema, metric expectations, and trainer smoke coverage.\n\n## CLI\u002Fconfig workflow\n\n- The `torchgeo` console script is declared as `torchgeo.main:main`.\n- Prefer JSONArgParse\u002FLightning config files for reproducible training. Explicitly name the datamodule, task\u002Fmodel, trainer, optimizer\u002Fscheduler parameters, and paths.\n- For smoke checks, use small fake fixtures and `max_epochs=1`, small batch sizes, and no checkpoint downloads.\n\n## Batch and mask handling\n\n- Images are typically tensors shaped `(B, C, H, W)`; some spatiotemporal paths use `(B, T, C, H, W)` and may flatten or route to spatiotemporal tasks.\n- Segmentation masks may require dtype conversion for Kornia and restoration before loss\u002Fmetrics.\n- Detection and instance segmentation batches can contain variable-length boxes\u002Fmasks; use the datamodule utility collate function used by the tests.\n\n## Read next\n\n- [reference](references\u002Fdatamodules-and-tasks.md) for task-specific examples and verification candidates.\n- Root [troubleshooting](..\u002F..\u002Freferences\u002Ftroubleshooting.md) for Lightning, Kornia, and optional dependency failures.\n","# TorchGeo datamodules and tasks\n\nUse this sub-skill when wiring TorchGeo into Lightning training, validation, testing, prediction, CLI configs, or task modules.\n\n## DataModule operating facts\n\n- `BaseDataModule` wraps a dataset class, `batch_size`, `num_workers`, split datasets, split dataloaders, and Kornia augmentation hooks.\n- `prepare_data()` may download data when `download=True`; it must not create split state. `setup(stage)` creates datasets\u002Fsamplers and is called on every process.\n- `GeoDataModule` is for `GeoDataset` classes and sets `collate_fn = stack_samples`. It manages `RandomPatchSampler`, `GriddedPatchSampler`, and batch samplers for split-specific patch extraction.\n- `NonGeoDataModule` is for integer-indexed datasets and ordinary PyTorch dataloader shuffle\u002Fsplit semantics.\n- `on_after_batch_transfer()` applies the split-specific augmentation after the batch reaches the device. This is where Kornia GPU augmentations belong.\n\n## Task classes\n\nTorchGeo task modules are Lightning modules for common geospatial ML problems:\n\n- `Classification`: timm-backed image classification with `weights`, `in_channels`, task type, class\u002Flabel counts, losses, and optional frozen backbone.\n- `SemanticSegmentation`: SMP or TorchGeo FCN segmentation with `model`, `backbone`, `weights`, `in_channels`, multiclass\u002Fmultilabel\u002Fbinary settings, losses, and freeze options.\n- Other task modules cover change detection, detection, instance segmentation, regression, temporal\u002Fspatiotemporal tasks, MAE, MoCo, BYOL, SimCLR, and IO benchmark workflows.\n\nBefore changing a task, inspect its corresponding `tests\u002Ftasks\u002Ftest_\u003Ctask>.py` for constructor matrix, batch schema, metric expectations, and trainer smoke coverage.\n\n## CLI\u002Fconfig workflow\n\n- The `torchgeo` console script is declared as `torchgeo.main:main`.\n- Prefer JSONArgParse\u002FLightning config files for reproducible training. Explicitly name the datamodule, task\u002Fmodel, trainer, optimizer\u002Fscheduler parameters, and paths.\n- For smoke checks, use small fake fixtures and `max_epochs=1`, small batch sizes, and no checkpoint downloads.\n\n## Batch and mask handling\n\n- Images are typically tensors shaped `(B, C, H, W)`; some spatiotemporal paths use `(B, T, C, H, W)` and may flatten or route to spatiotemporal tasks.\n- Segmentation masks may require dtype conversion for Kornia and restoration before loss\u002Fmetrics.\n- Detection and instance segmentation batches can contain variable-length boxes\u002Fmasks; use the datamodule utility collate function used by the tests.\n\n## Read next\n\n- [reference](references\u002Fdatamodules-and-tasks.md) for task-specific examples and verification candidates.\n- Root [troubleshooting](..\u002F..\u002Freferences\u002Ftroubleshooting.md) for Lightning, Kornia, and optional dependency 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\u002Ftorchgeo\u002Fsub-skills\u002Fdatamodules-and-tasks",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 datamodules-and-tasks",{"tool":39,"command":40},"bunx","bunx skills add VectorSpaceLab\u002FAREX-Skill --skill datamodules-and-tasks",{"tool":42,"command":43},"pnpm","pnpm dlx skills add VectorSpaceLab\u002FAREX-Skill --skill datamodules-and-tasks","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Ftorchgeo\u002Fsub-skills\u002Fdatamodules-and-tasks\u002FSKILL.md",[46],{"path":47,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":50,"sourceHtmlUrl":51},"references\u002Fdatamodules-and-tasks.md","REFERENCE","参考文档",2475,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Ftorchgeo\u002Fsub-skills\u002Fdatamodules-and-tasks\u002Freferences\u002Fdatamodules-and-tasks.md",[53,61,68,75,82,89],{"skillNo":54,"name":55,"nameZh":6,"description":56,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":57,"licenseFileUrl":6,"sourceHtmlUrl":58,"categories":59,"keywords":60,"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":62,"name":63,"nameZh":6,"description":64,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":57,"licenseFileUrl":6,"sourceHtmlUrl":65,"categories":66,"keywords":67,"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":69,"name":70,"nameZh":6,"description":71,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":57,"licenseFileUrl":6,"sourceHtmlUrl":72,"categories":73,"keywords":74,"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":76,"name":77,"nameZh":6,"description":78,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":57,"licenseFileUrl":6,"sourceHtmlUrl":79,"categories":80,"keywords":81,"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":83,"name":84,"nameZh":6,"description":85,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":57,"licenseFileUrl":6,"sourceHtmlUrl":86,"categories":87,"keywords":88,"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":90,"name":91,"nameZh":6,"description":92,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":57,"licenseFileUrl":6,"sourceHtmlUrl":93,"categories":94,"keywords":95,"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",[],[]]