[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7979S1NMAESXWABJ72B3NET":3,"skill-same:01M7979S1NMAESXWABJ72B3NET":67},{"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},"01M7979S1NMAESXWABJ72B3NET","transform-ops",null,"Use VoxelMorph PyTorch tensor transform operations for affine, displacement, coordinate, warping, integration, resizing, composition, random transform, and module wrapper 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\u002Fvoxelmorph\u002Fsub-skills\u002Ftransform-ops\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: transform-ops\ndescription: \"Use VoxelMorph PyTorch tensor transform operations for affine,\n  displacement, coordinate, warping, integration, resizing, composition, random\n  transform, and module wrapper workflows.\"\ndisable-model-invocation: true\nmetadata:\n  disco-role: operating\nlicense: Apache 2.0\n---\n\n# VoxelMorph transform operations\n\nUse this sub-skill when the task is about VoxelMorph's PyTorch tensor transform primitives, not about model training or data generation. It covers dense displacement fields, affine transforms, coordinate normalization, spatial warping, scaling-and-squaring integration, displacement resizing\u002Fcomposition, random transform generation, and the module wrappers around these operations.\n\n## Use this for\n\n- Building 2D\u002F3D affine matrices with `voxelmorph.params_to_affine()` or `voxelmorph.random_affine()`.\n- Converting affine matrices to dense displacement fields with `voxelmorph.affine_to_disp()`.\n- Converting among displacement fields, absolute transformation fields, and normalized sampling coordinates with `disp_to_trf()`, `trf_to_disp()`, and `disp_to_coords()`.\n- Warping tensors with `voxelmorph.spatial_transform()` or `voxelmorph.nn.functional.spatial_transform()`.\n- Integrating stationary velocity fields using scaling and squaring with `integrate_disp()` or `IntegrateVelocityField`.\n- Resizing\u002Frescaling dense displacement fields with `resize_disp()` or `ResizeDisplacementField`.\n- Composing affine and dense transforms with `compose()`.\n- Creating synthetic affine, displacement, or mixed random transforms for tests and augmentation-like utilities.\n- Using `SpatialTransformer`, `IntegrateVelocityField`, and `ResizeDisplacementField` inside PyTorch modules.\n\n## Do not use this for\n\n- Pairwise-registration model construction, loss selection, optimizer setup, training loops, checkpoints, or inference workflows; use the pairwise-registration sub-skill instead.\n- Volume loading, file I\u002FO, batching, generator design, scan pair sampling, or data augmentation pipelines around datasets; use the data-generators sub-skill instead.\n- TensorFlow-era VoxelMorph APIs. This sub-skill describes the PyTorch branch behavior.\n\n## First routing decisions\n\n1. **Choose the API layer.** Use top-level `voxelmorph` functions for unbatched fields, affine utilities, and shape-agnostic transforms. Use `voxelmorph.nn.functional` when tensors are already in neural-network format `(B, C, *spatial)` for images and `(B, ndim, *spatial)` for fields. Use module wrappers when the operation belongs in an `nn.Module` graph.\n2. **Normalize shape language.** Dense fields are channels-first: `(ndim, *spatial)` or `(B, ndim, *spatial)`. Image tensors use `non_spatial_dims` to identify leading channel\u002Fbatch dimensions. Meshgrids are `(ndim, *spatial)` with `ij` indexing.\n3. **Remember backward warping.** `spatial_transform()` samples the source image at `x + disp(x)`, so a positive displacement samples from a larger coordinate and can make visible image content move in the opposite direction.\n4. **Treat `coords_to_disp()` as unavailable.** It exists in both public layers but currently raises `NotImplementedError`; do not plan workflows that require the inverse normalized-coordinate conversion.\n5. **Keep model and data responsibilities out.** If the task starts asking for training, VxmPairwise, losses, volume readers, or generators, stop routing through this sub-skill.\n\n## Reference map\n\n- [API reference](references\u002Fapi-reference.md): signatures, shape conventions, return types, and gotchas for `voxelmorph.functional`, `voxelmorph.nn.functional`, and module wrappers.\n- [Workflows](references\u002Fworkflows.md): copied\u002Fadapted runnable patterns for affine-to-displacement, warping, integration, resizing, composition, random transforms, and module use.\n- [Troubleshooting](references\u002Ftroubleshooting.md): common assertion errors, axis\u002Fsign mistakes, interpolation issues, unavailable inverse conversion, random seeding, and composition limitations.\n- [Smoke script](scripts\u002Ftransform_ops_smoke.py): deterministic CPU\u002FGPU-safe checks for core transform operations; run with `python scripts\u002Ftransform_ops_smoke.py --help` from this sub-skill directory or with the script path from any project that has `voxelmorph` importable.\n","# VoxelMorph transform operations\n\nUse this sub-skill when the task is about VoxelMorph's PyTorch tensor transform primitives, not about model training or data generation. It covers dense displacement fields, affine transforms, coordinate normalization, spatial warping, scaling-and-squaring integration, displacement resizing\u002Fcomposition, random transform generation, and the module wrappers around these operations.\n\n## Use this for\n\n- Building 2D\u002F3D affine matrices with `voxelmorph.params_to_affine()` or `voxelmorph.random_affine()`.\n- Converting affine matrices to dense displacement fields with `voxelmorph.affine_to_disp()`.\n- Converting among displacement fields, absolute transformation fields, and normalized sampling coordinates with `disp_to_trf()`, `trf_to_disp()`, and `disp_to_coords()`.\n- Warping tensors with `voxelmorph.spatial_transform()` or `voxelmorph.nn.functional.spatial_transform()`.\n- Integrating stationary velocity fields using scaling and squaring with `integrate_disp()` or `IntegrateVelocityField`.\n- Resizing\u002Frescaling dense displacement fields with `resize_disp()` or `ResizeDisplacementField`.\n- Composing affine and dense transforms with `compose()`.\n- Creating synthetic affine, displacement, or mixed random transforms for tests and augmentation-like utilities.\n- Using `SpatialTransformer`, `IntegrateVelocityField`, and `ResizeDisplacementField` inside PyTorch modules.\n\n## Do not use this for\n\n- Pairwise-registration model construction, loss selection, optimizer setup, training loops, checkpoints, or inference workflows; use the pairwise-registration sub-skill instead.\n- Volume loading, file I\u002FO, batching, generator design, scan pair sampling, or data augmentation pipelines around datasets; use the data-generators sub-skill instead.\n- TensorFlow-era VoxelMorph APIs. This sub-skill describes the PyTorch branch behavior.\n\n## First routing decisions\n\n1. **Choose the API layer.** Use top-level `voxelmorph` functions for unbatched fields, affine utilities, and shape-agnostic transforms. Use `voxelmorph.nn.functional` when tensors are already in neural-network format `(B, C, *spatial)` for images and `(B, ndim, *spatial)` for fields. Use module wrappers when the operation belongs in an `nn.Module` graph.\n2. **Normalize shape language.** Dense fields are channels-first: `(ndim, *spatial)` or `(B, ndim, *spatial)`. Image tensors use `non_spatial_dims` to identify leading channel\u002Fbatch dimensions. Meshgrids are `(ndim, *spatial)` with `ij` indexing.\n3. **Remember backward warping.** `spatial_transform()` samples the source image at `x + disp(x)`, so a positive displacement samples from a larger coordinate and can make visible image content move in the opposite direction.\n4. **Treat `coords_to_disp()` as unavailable.** It exists in both public layers but currently raises `NotImplementedError`; do not plan workflows that require the inverse normalized-coordinate conversion.\n5. **Keep model and data responsibilities out.** If the task starts asking for training, VxmPairwise, losses, volume readers, or generators, stop routing through this sub-skill.\n\n## Reference map\n\n- [API reference](references\u002Fapi-reference.md): signatures, shape conventions, return types, and gotchas for `voxelmorph.functional`, `voxelmorph.nn.functional`, and module wrappers.\n- [Workflows](references\u002Fworkflows.md): copied\u002Fadapted runnable patterns for affine-to-displacement, warping, integration, resizing, composition, random transforms, and module use.\n- [Troubleshooting](references\u002Ftroubleshooting.md): common assertion errors, axis\u002Fsign mistakes, interpolation issues, unavailable inverse conversion, random seeding, and composition limitations.\n- [Smoke script](scripts\u002Ftransform_ops_smoke.py): deterministic CPU\u002FGPU-safe checks for core transform operations; run with `python scripts\u002Ftransform_ops_smoke.py --help` from this sub-skill directory or with the script path from any project that has `voxelmorph` importable.\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\u002Fvoxelmorph\u002Fsub-skills\u002Ftransform-ops",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 transform-ops",{"tool":40,"command":41},"bunx","bunx skills add VectorSpaceLab\u002FAREX-Skill --skill transform-ops",{"tool":43,"command":44},"pnpm","pnpm dlx skills add VectorSpaceLab\u002FAREX-Skill --skill transform-ops","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fvoxelmorph\u002Fsub-skills\u002Ftransform-ops\u002FSKILL.md",[47,53,57,61],{"path":48,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":51,"sourceHtmlUrl":52},"references\u002Fapi-reference.md","REFERENCE","参考文档",13489,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fvoxelmorph\u002Fsub-skills\u002Ftransform-ops\u002Freferences\u002Fapi-reference.md",{"path":54,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":55,"sourceHtmlUrl":56},"references\u002Ftroubleshooting.md",7061,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fvoxelmorph\u002Fsub-skills\u002Ftransform-ops\u002Freferences\u002Ftroubleshooting.md",{"path":58,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":59,"sourceHtmlUrl":60},"references\u002Fworkflows.md",6691,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fvoxelmorph\u002Fsub-skills\u002Ftransform-ops\u002Freferences\u002Fworkflows.md",{"path":62,"resourceType":63,"resourceTypeDesc":64,"sizeBytes":65,"sourceHtmlUrl":66},"scripts\u002Ftransform_ops_smoke.py","SCRIPT","脚本",8572,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fvoxelmorph\u002Fsub-skills\u002Ftransform-ops\u002Fscripts\u002Ftransform_ops_smoke.py",[68,75,82,89,96,103],{"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":14,"licenseFileUrl":6,"sourceHtmlUrl":72,"categories":73,"keywords":74,"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":76,"name":77,"nameZh":6,"description":78,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":79,"categories":80,"keywords":81,"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":83,"name":84,"nameZh":6,"description":85,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":86,"categories":87,"keywords":88,"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":90,"name":91,"nameZh":6,"description":92,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":93,"categories":94,"keywords":95,"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":97,"name":98,"nameZh":6,"description":99,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":100,"categories":101,"keywords":102,"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":104,"name":105,"nameZh":6,"description":106,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":107,"categories":108,"keywords":109,"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",[],[]]