[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7FCRG3KXRP7KZTHC0PAMWMC":3,"skill-same:01M7FCRG3KXRP7KZTHC0PAMWMC":63},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":18,"publishedAt":19,"linkOnly":20,"markdownContent":21,"markdownBody":22,"frontMatter":23,"refName":28,"commitSha":29,"skillDirectory":30,"forksCount":31,"repositoryHtmlUrl":32,"repositoryHomepageUrl":6,"ownerHtmlUrl":33,"ownerBlogUrl":6,"ownerDescription":34,"updatedAt":35,"install":36,"resources":48},"01M7FCRG3KXRP7KZTHC0PAMWMC","patch-workflows",null,"Design TorchIO patch-based training queues and dense patch inference loops with samplers, SubjectsLoader batches, PatchLocation metadata, and PatchAggregator.","lza6\u002FArmor-piercing-system-prompts","lza6","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F128880206?v=4","01M7EB3K7WPX5T28EJ2NVHE6EG","Chinese-tingfeng","01M7EB3K97E1KQ16QWKJMFNGX8",11,"Apache-2.0","https:\u002F\u002Fgithub.com\u002Flza6\u002FArmor-piercing-system-prompts\u002Fblob\u002Fmain\u002F06-前端与设计\u002Ftorchio\u002Fsub-skills\u002Fpatch-workflows\u002FSKILL.md",[],[],"2026-11-16 10:05:34",false,"---\nname: patch-workflows\ndescription: \"Design TorchIO patch-based training queues and dense patch\n  inference loops with samplers, SubjectsLoader batches, PatchLocation metadata,\n  and PatchAggregator.\"\ndisable-model-invocation: true\nmetadata:\n  disco-role: operating\nlicense: Apache 2.0\n---\n\n# Patch Workflows\n\nUse this sub-skill when an agent needs to train or infer on large 3D images by sampling patches instead of processing whole volumes. It covers `GridSampler`, `UniformSampler`, `WeightedSampler`, `LabelSampler`, `Queue`, `SubjectsLoader`, `ImagesLoader`, `PatchLocation`, and `PatchAggregator`.\n\n## Route by Goal\n\n- Need dense whole-volume inference: use `GridSampler` plus `PatchAggregator`; see [dense inference](references\u002Fworkflows.md#dense-inference-with-grid-sampler-and-patchaggregator).\n- Need stochastic training patches: use `UniformSampler`, `WeightedSampler`, or `LabelSampler`, optionally buffered by `Queue`; see [training queues](references\u002Fworkflows.md#training-with-queue-and-random-samplers).\n- Need to choose sampler or batch access patterns: see [API reference](references\u002Fapi-reference.md).\n- Hit shape, overlap, probability map, or metadata errors: see [troubleshooting](references\u002Ftroubleshooting.md).\n- Want a runnable synthetic sanity check: run `python scripts\u002Fpatch_workflow_smoke.py --help`, then `python scripts\u002Fpatch_workflow_smoke.py` from this sub-skill directory.\n\n## Current API Reminders\n\n- Construct in-memory images as `tio.ScalarImage(tensor)` and labels as `tio.LabelMap(tensor)`; do not use `source=tensor`.\n- Image tensors are 4D `(C, I, J, K)` and loader batches are 5D `(B, C, I, J, K)`.\n- `SubjectsLoader` returns a `SubjectsBatch`: named images are accessed as `batch.t1.data` or `batch[\"t1\"].data`.\n- Patch locations are stored on individual patch subjects as `patch.patch_location`; after batching, use `batch.metadata[\"patch_location\"]`.\n- `PatchAggregator.add_batch()` expects model outputs shaped `(B, C, I, J, K)` and a list of one `PatchLocation` per batch item.\n\n## Minimal Dense Inference Pattern\n\n```python\nsampler = tio.GridSampler(subject, patch_size=(64, 64, 64), patch_overlap=16)\nloader = tio.SubjectsLoader(sampler, batch_size=4)\naggregator = tio.PatchAggregator(\n    spatial_shape=subject.spatial_shape,\n    overlap_mode=\"hann\",\n    patch_overlap=16,\n)\n\nfor batch in loader:\n    logits = model(batch.t1.data)\n    locations = batch.metadata[\"patch_location\"]\n    aggregator.add_batch(logits, locations)\n\nvolume_logits = aggregator.get_output()\n```\n\n## Minimal Training Queue Pattern\n\n```python\nsampler = tio.UniformSampler(subjects[0], patch_size=(64, 64, 64))\nqueue = tio.Queue(\n    subjects,\n    patch_sampler=sampler,\n    max_length=300,\n    patches_per_volume=10,\n    num_workers=4,\n    transform=train_transform,\n)\nloader = tio.SubjectsLoader(queue, batch_size=16)\n\nfor batch in loader:\n    inputs = batch.t1.data\n    targets = batch.seg.data\n    loss = criterion(model(inputs), targets)\n    loss.backward()\n```\n\n## Boundaries\n\n- For creating `Subject`, `ScalarImage`, and `LabelMap` objects, use the parent data-model guidance.\n- For choosing augmentations or preprocessing transforms, use the transforms sub-skill.\n- For command-line operations, use the CLI and I\u002FO sub-skill.\n","# Patch Workflows\n\nUse this sub-skill when an agent needs to train or infer on large 3D images by sampling patches instead of processing whole volumes. It covers `GridSampler`, `UniformSampler`, `WeightedSampler`, `LabelSampler`, `Queue`, `SubjectsLoader`, `ImagesLoader`, `PatchLocation`, and `PatchAggregator`.\n\n## Route by Goal\n\n- Need dense whole-volume inference: use `GridSampler` plus `PatchAggregator`; see [dense inference](references\u002Fworkflows.md#dense-inference-with-grid-sampler-and-patchaggregator).\n- Need stochastic training patches: use `UniformSampler`, `WeightedSampler`, or `LabelSampler`, optionally buffered by `Queue`; see [training queues](references\u002Fworkflows.md#training-with-queue-and-random-samplers).\n- Need to choose sampler or batch access patterns: see [API reference](references\u002Fapi-reference.md).\n- Hit shape, overlap, probability map, or metadata errors: see [troubleshooting](references\u002Ftroubleshooting.md).\n- Want a runnable synthetic sanity check: run `python scripts\u002Fpatch_workflow_smoke.py --help`, then `python scripts\u002Fpatch_workflow_smoke.py` from this sub-skill directory.\n\n## Current API Reminders\n\n- Construct in-memory images as `tio.ScalarImage(tensor)` and labels as `tio.LabelMap(tensor)`; do not use `source=tensor`.\n- Image tensors are 4D `(C, I, J, K)` and loader batches are 5D `(B, C, I, J, K)`.\n- `SubjectsLoader` returns a `SubjectsBatch`: named images are accessed as `batch.t1.data` or `batch[\"t1\"].data`.\n- Patch locations are stored on individual patch subjects as `patch.patch_location`; after batching, use `batch.metadata[\"patch_location\"]`.\n- `PatchAggregator.add_batch()` expects model outputs shaped `(B, C, I, J, K)` and a list of one `PatchLocation` per batch item.\n\n## Minimal Dense Inference Pattern\n\n```python\nsampler = tio.GridSampler(subject, patch_size=(64, 64, 64), patch_overlap=16)\nloader = tio.SubjectsLoader(sampler, batch_size=4)\naggregator = tio.PatchAggregator(\n    spatial_shape=subject.spatial_shape,\n    overlap_mode=\"hann\",\n    patch_overlap=16,\n)\n\nfor batch in loader:\n    logits = model(batch.t1.data)\n    locations = batch.metadata[\"patch_location\"]\n    aggregator.add_batch(logits, locations)\n\nvolume_logits = aggregator.get_output()\n```\n\n## Minimal Training Queue Pattern\n\n```python\nsampler = tio.UniformSampler(subjects[0], patch_size=(64, 64, 64))\nqueue = tio.Queue(\n    subjects,\n    patch_sampler=sampler,\n    max_length=300,\n    patches_per_volume=10,\n    num_workers=4,\n    transform=train_transform,\n)\nloader = tio.SubjectsLoader(queue, batch_size=16)\n\nfor batch in loader:\n    inputs = batch.t1.data\n    targets = batch.seg.data\n    loss = criterion(model(inputs), targets)\n    loss.backward()\n```\n\n## Boundaries\n\n- For creating `Subject`, `ScalarImage`, and `LabelMap` objects, use the parent data-model guidance.\n- For choosing augmentations or preprocessing transforms, use the transforms sub-skill.\n- For command-line operations, use the CLI and I\u002FO sub-skill.\n",{"name":5,"license":24,"metadata":25,"description":7,"disable-model-invocation":27},"Apache 2.0",{"disco-role":26},"operating",true,"main","d69096e239f2a89aabc2cd078c47664a9d3b1e11","06-前端与设计\u002Ftorchio\u002Fsub-skills\u002Fpatch-workflows",3,"https:\u002F\u002Fgithub.com\u002Flza6\u002FArmor-piercing-system-prompts","https:\u002F\u002Fgithub.com\u002Flza6","个人vx：Tf00798\r\n个人✈：@TFGY999","2026-10-08 14:29:22",{"source":8,"skillName":5,"commands":37,"prompt":47,"zipAvailable":27},[38,41,44],{"tool":39,"command":40},"npx","npx skills add lza6\u002FArmor-piercing-system-prompts --skill patch-workflows",{"tool":42,"command":43},"bunx","bunx skills add lza6\u002FArmor-piercing-system-prompts --skill patch-workflows",{"tool":45,"command":46},"pnpm","pnpm dlx skills add lza6\u002FArmor-piercing-system-prompts --skill patch-workflows","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Flza6\u002FArmor-piercing-system-prompts\u002Fblob\u002Fmain\u002F06-前端与设计\u002Ftorchio\u002Fsub-skills\u002Fpatch-workflows\u002FSKILL.md",[49,55,59],{"path":50,"resourceType":51,"resourceTypeDesc":52,"sizeBytes":53,"sourceHtmlUrl":54},"references\u002Fapi-reference.md","REFERENCE","参考文档",6054,"https:\u002F\u002Fgithub.com\u002Flza6\u002FArmor-piercing-system-prompts\u002Fblob\u002Fmain\u002F06-前端与设计\u002Ftorchio\u002Fsub-skills\u002Fpatch-workflows\u002Freferences\u002Fapi-reference.md",{"path":56,"resourceType":51,"resourceTypeDesc":52,"sizeBytes":57,"sourceHtmlUrl":58},"references\u002Ftroubleshooting.md",5802,"https:\u002F\u002Fgithub.com\u002Flza6\u002FArmor-piercing-system-prompts\u002Fblob\u002Fmain\u002F06-前端与设计\u002Ftorchio\u002Fsub-skills\u002Fpatch-workflows\u002Freferences\u002Ftroubleshooting.md",{"path":60,"resourceType":51,"resourceTypeDesc":52,"sizeBytes":61,"sourceHtmlUrl":62},"references\u002Fworkflows.md",5693,"https:\u002F\u002Fgithub.com\u002Flza6\u002FArmor-piercing-system-prompts\u002Fblob\u002Fmain\u002F06-前端与设计\u002Ftorchio\u002Fsub-skills\u002Fpatch-workflows\u002Freferences\u002Fworkflows.md",[64,72,77,84,91,98],{"skillNo":65,"name":66,"nameZh":6,"description":67,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":68,"licenseFileUrl":6,"sourceHtmlUrl":69,"categories":70,"keywords":71,"publishedAt":19},"01M7FFG485KY83TD6HSC4YPCMC","obsidian-clipper-template-creator","Guide for creating templates for the Obsidian Web Clipper. Use when you want to create a new clipping template, understand available variables, or format clipped content.","MIT","https:\u002F\u002Fgithub.com\u002Flza6\u002FArmor-piercing-system-prompts\u002Fblob\u002Fmain\u002F07-后端与架构\u002Fobsidian-clipper-template-creator\u002FSKILL.md",[],[],{"skillNo":73,"name":66,"nameZh":6,"description":67,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":68,"licenseFileUrl":6,"sourceHtmlUrl":74,"categories":75,"keywords":76,"publishedAt":19},"01M7FFG0T6AZ0STEJPSB3CX84K","https:\u002F\u002Fgithub.com\u002Flza6\u002FArmor-piercing-system-prompts\u002Fblob\u002Fmain\u002F07-后端与架构\u002Fobsidian-clipper-template-creator-antigravity-awesome-skills-main\u002FSKILL.md",[],[],{"skillNo":78,"name":79,"nameZh":6,"description":80,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":68,"licenseFileUrl":6,"sourceHtmlUrl":81,"categories":82,"keywords":83,"publishedAt":19},"01M7FFFY0QEM14SKWPDYGMXM2P","python-api","Programmatic Object-Detection-Metrics API guidance for BoundingBox, BoundingBoxes, Evaluator, enums, coordinate conversion, AP\u002FmAP semantics, and source-style import diagnostics.","https:\u002F\u002Fgithub.com\u002Flza6\u002FArmor-piercing-system-prompts\u002Fblob\u002Fmain\u002F07-后端与架构\u002Fobject-detection-metrics\u002Fsub-skills\u002Fpython-api\u002FSKILL.md",[],[],{"skillNo":85,"name":86,"nameZh":6,"description":87,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":68,"licenseFileUrl":6,"sourceHtmlUrl":88,"categories":89,"keywords":90,"publishedAt":19},"01M7FFFVZQZ2GJ0C0TGP7JHD20","file-evaluation","Folder-based PASCAL VOC AP\u002FmAP evaluation from ground-truth and detection text files.","https:\u002F\u002Fgithub.com\u002Flza6\u002FArmor-piercing-system-prompts\u002Fblob\u002Fmain\u002F07-后端与架构\u002Fobject-detection-metrics\u002Fsub-skills\u002Ffile-evaluation\u002FSKILL.md",[],[],{"skillNo":92,"name":93,"nameZh":6,"description":94,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":68,"licenseFileUrl":6,"sourceHtmlUrl":95,"categories":96,"keywords":97,"publishedAt":19},"01M7FFFSFNYSE77SJ6PE1CK2DA","object-detection-metrics","Guides legacy Object-Detection-Metrics PASCAL VOC AP\u002FmAP evaluation from detection text folders and direct Python API classes.","https:\u002F\u002Fgithub.com\u002Flza6\u002FArmor-piercing-system-prompts\u002Fblob\u002Fmain\u002F07-后端与架构\u002Fobject-detection-metrics\u002FSKILL.md",[],[],{"skillNo":99,"name":100,"nameZh":6,"description":101,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":68,"licenseFileUrl":6,"sourceHtmlUrl":102,"categories":103,"keywords":104,"publishedAt":19},"01M7FFFP0GCK1RVXBCPQ8APBMT","nx-workspace-patterns","Configure and optimize Nx monorepo workspaces. Use when setting up Nx, configuring project boundaries, optimizing build caching, or implementing affected commands.","https:\u002F\u002Fgithub.com\u002Flza6\u002FArmor-piercing-system-prompts\u002Fblob\u002Fmain\u002F07-后端与架构\u002Fnx-workspace-patterns\u002FSKILL.md",[],[]]