[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M78VMFBSGEE9V52RF82SS11M":3,"skill-same:01M78VMFBSGEE9V52RF82SS11M":82},{"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},"01M78VMFBSGEE9V52RF82SS11M","inference-evaluation",null,"Run and troubleshoot DeepDanbooru image-tag inference through the project or explicit model\u002Ftag APIs, including preprocessing, folder traversal, thresholds, and text outputs.","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\u002Fdeep-danbooru\u002Fsub-skills\u002Finference-evaluation\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: inference-evaluation\ndescription: \"Run and troubleshoot DeepDanbooru image-tag inference through the\n  project or explicit model\u002Ftag APIs, including preprocessing, folder traversal,\n  thresholds, and text outputs.\"\ndisable-model-invocation: true\nmetadata:\n  disco-role: operating\nlicense: MIT\n---\n\n# Inference and evaluation\n\nUse this sub-skill when a trained DeepDanbooru model must turn one or more\nimages into tags. It covers the `evaluate` and `evaluate-project` CLI commands,\nthe corresponding Python functions, image preprocessing, recursive folder\nselection, thresholds, and optional `.txt` sidecars.\n\nThis is an inference skill, not a training or dataset-preparation procedure.\nFor a missing or malformed project, tag file, or dataset, use\n[project-data-setup](..\u002Fproject-data-setup\u002FSKILL.md) first. For model conversion\nor experimental Grad-CAM output, use\n[post-training-tools](..\u002Fpost-training-tools\u002FSKILL.md). The bundled smoke\nscripts are tiny local checks; they do not download weights, contact a network,\nor read the original checkout.\n\n## Choose the entry point\n\n| Situation | Entry point | Important constraint |\n|---|---|---|\n| Use a project to select both model and tags | `deepdanbooru evaluate TARGET --project-path PROJECT` | Project must contain `project.json`, `tags.txt`, and its selected model artifact. |\n| Use an explicit saved model and tag file | `deepdanbooru evaluate TARGET --model-path MODEL --tags-path TAGS` | Without `--project-path`, **both** explicit paths are required. |\n| Override only one project component | `evaluate` with project plus `--model-path` or `--tags-path` | A direct model or tags path takes precedence for that component. |\n| Evaluate a file or recursively every recognized image below a folder | `deepdanbooru evaluate TARGET --project-path PROJECT --allow-folder` | This is the supported 1.0.0 project-folder path. Native `evaluate-project` is not recommended because its loader calls a missing symbol; see below. |\n| Call inference from Python | `deepdanbooru.commands.evaluate_image` or `evaluate` | Match the model's NHWC input dimensions and tag-output dimension. |\n\nThe `evaluate` function accepts a folder only when `allow_folder=True`. The\nClick CLI defaults that flag off, so passing a directory without `--allow-folder`\nis a common mistake. In the exact 1.0.0 source revision, native\n`evaluate-project` calls the missing `dd.data.load_tags_from_project` symbol and\nraises `AttributeError` before inference. Use `evaluate` with `--project-path`\nand `--allow-folder` for production. The bundled project smoke helper exercises\nthat supported fallback; it does not claim a native `evaluate-project` run.\n\n## CPU-first prerequisites\n\nThe verified environment is Python 3.11 with TensorFlow 2.21.0 and no CUDA\nlibraries in the verification prefix. CPU correctness is required. By default,\n`evaluate` sets `CUDA_VISIBLE_DEVICES=-1` before loading the model; do not remove\nthat safeguard for a CPU run. `--allow-gpu` only leaves device selection\nunchanged. It is an opt-in switch, **not evidence that GPU support is ready**;\nGPU readiness was not verified for this repository.\n\nBefore a real run, confirm:\n\n1. `deepdanbooru --help` and `deepdanbooru evaluate --help` work in the chosen\n   environment.\n2. The model is loadable and has an image input shape with height at index 1 and\n   width at index 2, for example `(None, 299, 299, 3)`.\n3. The tag file is newline-separated and is the exact vocabulary used for the\n   model's output units.\n4. Target paths exist and contain decodable images. `--save-txt` writes beside\n   each image and the native helper can overwrite an existing sidecar. For a\n   guarded skill-layer write, use `scripts\u002Fsave_txt_guard.py`: it rejects an\n   empty selected-tag list, rejects a sibling directory even with overwrite\n   enabled, and requires explicit overwrite permission for regular files.\n\n## Fast, inspectable workflow\n\n1. Start with one known image and `--threshold 0.5`, without `--save-txt`.\n2. Choose the project route or provide both explicit model and tags paths. Use\n   `--verbose` when diagnosing path selection or load failures.\n3. Inspect the printed scores and tag order. If no tag is printed, rerun with a\n   lower diagnostic threshold such as `0.0`; do not immediately assume the\n   image decoder or model failed.\n4. Add `--allow-folder` and a narrow `--folder-filters` only after one-file\n   inference works. Confirm the discovered set and natural ordering.\n5. Enable `--save-txt` only after output is useful, and verify each sidecar.\n   The native path writes selected tag names only and replaces existing files;\n   use the guard helper when empty output or replacement must be refused.\n6. Use the deterministic helpers when the environment or a fixture is in doubt:\n   `python scripts\u002Fimage_preprocess_smoke.py`,\n   `python scripts\u002Fdummy_evaluate_smoke.py`, and\n   `python scripts\u002Fevaluate_project_smoke.py`.\n\nRead the focused contracts before changing a command:\n\n- [CLI reference](references\u002Fcli-reference.md)\n- [Python API reference](references\u002Fapi-reference.md)\n- [Image inputs and tags](references\u002Fimage-inputs-and-tags.md)\n- [Troubleshooting](references\u002Ftroubleshooting.md)\n","# Inference and evaluation\n\nUse this sub-skill when a trained DeepDanbooru model must turn one or more\nimages into tags. It covers the `evaluate` and `evaluate-project` CLI commands,\nthe corresponding Python functions, image preprocessing, recursive folder\nselection, thresholds, and optional `.txt` sidecars.\n\nThis is an inference skill, not a training or dataset-preparation procedure.\nFor a missing or malformed project, tag file, or dataset, use\n[project-data-setup](..\u002Fproject-data-setup\u002FSKILL.md) first. For model conversion\nor experimental Grad-CAM output, use\n[post-training-tools](..\u002Fpost-training-tools\u002FSKILL.md). The bundled smoke\nscripts are tiny local checks; they do not download weights, contact a network,\nor read the original checkout.\n\n## Choose the entry point\n\n| Situation | Entry point | Important constraint |\n|---|---|---|\n| Use a project to select both model and tags | `deepdanbooru evaluate TARGET --project-path PROJECT` | Project must contain `project.json`, `tags.txt`, and its selected model artifact. |\n| Use an explicit saved model and tag file | `deepdanbooru evaluate TARGET --model-path MODEL --tags-path TAGS` | Without `--project-path`, **both** explicit paths are required. |\n| Override only one project component | `evaluate` with project plus `--model-path` or `--tags-path` | A direct model or tags path takes precedence for that component. |\n| Evaluate a file or recursively every recognized image below a folder | `deepdanbooru evaluate TARGET --project-path PROJECT --allow-folder` | This is the supported 1.0.0 project-folder path. Native `evaluate-project` is not recommended because its loader calls a missing symbol; see below. |\n| Call inference from Python | `deepdanbooru.commands.evaluate_image` or `evaluate` | Match the model's NHWC input dimensions and tag-output dimension. |\n\nThe `evaluate` function accepts a folder only when `allow_folder=True`. The\nClick CLI defaults that flag off, so passing a directory without `--allow-folder`\nis a common mistake. In the exact 1.0.0 source revision, native\n`evaluate-project` calls the missing `dd.data.load_tags_from_project` symbol and\nraises `AttributeError` before inference. Use `evaluate` with `--project-path`\nand `--allow-folder` for production. The bundled project smoke helper exercises\nthat supported fallback; it does not claim a native `evaluate-project` run.\n\n## CPU-first prerequisites\n\nThe verified environment is Python 3.11 with TensorFlow 2.21.0 and no CUDA\nlibraries in the verification prefix. CPU correctness is required. By default,\n`evaluate` sets `CUDA_VISIBLE_DEVICES=-1` before loading the model; do not remove\nthat safeguard for a CPU run. `--allow-gpu` only leaves device selection\nunchanged. It is an opt-in switch, **not evidence that GPU support is ready**;\nGPU readiness was not verified for this repository.\n\nBefore a real run, confirm:\n\n1. `deepdanbooru --help` and `deepdanbooru evaluate --help` work in the chosen\n   environment.\n2. The model is loadable and has an image input shape with height at index 1 and\n   width at index 2, for example `(None, 299, 299, 3)`.\n3. The tag file is newline-separated and is the exact vocabulary used for the\n   model's output units.\n4. Target paths exist and contain decodable images. `--save-txt` writes beside\n   each image and the native helper can overwrite an existing sidecar. For a\n   guarded skill-layer write, use `scripts\u002Fsave_txt_guard.py`: it rejects an\n   empty selected-tag list, rejects a sibling directory even with overwrite\n   enabled, and requires explicit overwrite permission for regular files.\n\n## Fast, inspectable workflow\n\n1. Start with one known image and `--threshold 0.5`, without `--save-txt`.\n2. Choose the project route or provide both explicit model and tags paths. Use\n   `--verbose` when diagnosing path selection or load failures.\n3. Inspect the printed scores and tag order. If no tag is printed, rerun with a\n   lower diagnostic threshold such as `0.0`; do not immediately assume the\n   image decoder or model failed.\n4. Add `--allow-folder` and a narrow `--folder-filters` only after one-file\n   inference works. Confirm the discovered set and natural ordering.\n5. Enable `--save-txt` only after output is useful, and verify each sidecar.\n   The native path writes selected tag names only and replaces existing files;\n   use the guard helper when empty output or replacement must be refused.\n6. Use the deterministic helpers when the environment or a fixture is in doubt:\n   `python scripts\u002Fimage_preprocess_smoke.py`,\n   `python scripts\u002Fdummy_evaluate_smoke.py`, and\n   `python scripts\u002Fevaluate_project_smoke.py`.\n\nRead the focused contracts before changing a command:\n\n- [CLI reference](references\u002Fcli-reference.md)\n- [Python API reference](references\u002Fapi-reference.md)\n- [Image inputs and tags](references\u002Fimage-inputs-and-tags.md)\n- [Troubleshooting](references\u002Ftroubleshooting.md)\n",{"name":5,"license":14,"metadata":23,"description":7,"disable-model-invocation":25},{"disco-role":24},"operating",true,"main","ac3fe1afa80fb9a09775ecfb2b6cc3ba850a2db6","skills\u002Frepositories\u002Frepo-skills\u002Fdeep-danbooru\u002Fsub-skills\u002Finference-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 inference-evaluation",{"tool":39,"command":40},"bunx","bunx skills add VectorSpaceLab\u002FAREX-Skill --skill inference-evaluation",{"tool":42,"command":43},"pnpm","pnpm dlx skills add VectorSpaceLab\u002FAREX-Skill --skill inference-evaluation","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fdeep-danbooru\u002Fsub-skills\u002Finference-evaluation\u002FSKILL.md",[46,52,56,60,64,70,74,78],{"path":47,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":50,"sourceHtmlUrl":51},"references\u002Fapi-reference.md","REFERENCE","参考文档",6012,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fdeep-danbooru\u002Fsub-skills\u002Finference-evaluation\u002Freferences\u002Fapi-reference.md",{"path":53,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":54,"sourceHtmlUrl":55},"references\u002Fcli-reference.md",4679,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fdeep-danbooru\u002Fsub-skills\u002Finference-evaluation\u002Freferences\u002Fcli-reference.md",{"path":57,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":58,"sourceHtmlUrl":59},"references\u002Fimage-inputs-and-tags.md",4420,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fdeep-danbooru\u002Fsub-skills\u002Finference-evaluation\u002Freferences\u002Fimage-inputs-and-tags.md",{"path":61,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":62,"sourceHtmlUrl":63},"references\u002Ftroubleshooting.md",7428,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fdeep-danbooru\u002Fsub-skills\u002Finference-evaluation\u002Freferences\u002Ftroubleshooting.md",{"path":65,"resourceType":66,"resourceTypeDesc":67,"sizeBytes":68,"sourceHtmlUrl":69},"scripts\u002Fdummy_evaluate_smoke.py","SCRIPT","脚本",1731,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fdeep-danbooru\u002Fsub-skills\u002Finference-evaluation\u002Fscripts\u002Fdummy_evaluate_smoke.py",{"path":71,"resourceType":66,"resourceTypeDesc":67,"sizeBytes":72,"sourceHtmlUrl":73},"scripts\u002Fevaluate_project_smoke.py",3454,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fdeep-danbooru\u002Fsub-skills\u002Finference-evaluation\u002Fscripts\u002Fevaluate_project_smoke.py",{"path":75,"resourceType":66,"resourceTypeDesc":67,"sizeBytes":76,"sourceHtmlUrl":77},"scripts\u002Fimage_preprocess_smoke.py",1974,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fdeep-danbooru\u002Fsub-skills\u002Finference-evaluation\u002Fscripts\u002Fimage_preprocess_smoke.py",{"path":79,"resourceType":66,"resourceTypeDesc":67,"sizeBytes":80,"sourceHtmlUrl":81},"scripts\u002Fsave_txt_guard.py",2398,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fdeep-danbooru\u002Fsub-skills\u002Finference-evaluation\u002Fscripts\u002Fsave_txt_guard.py",[83,91,98,105,112,119],{"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":87,"licenseFileUrl":6,"sourceHtmlUrl":88,"categories":89,"keywords":90,"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":92,"name":93,"nameZh":6,"description":94,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":87,"licenseFileUrl":6,"sourceHtmlUrl":95,"categories":96,"keywords":97,"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":99,"name":100,"nameZh":6,"description":101,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":87,"licenseFileUrl":6,"sourceHtmlUrl":102,"categories":103,"keywords":104,"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":106,"name":107,"nameZh":6,"description":108,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":87,"licenseFileUrl":6,"sourceHtmlUrl":109,"categories":110,"keywords":111,"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":113,"name":114,"nameZh":6,"description":115,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":87,"licenseFileUrl":6,"sourceHtmlUrl":116,"categories":117,"keywords":118,"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":120,"name":121,"nameZh":6,"description":122,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":87,"licenseFileUrl":6,"sourceHtmlUrl":123,"categories":124,"keywords":125,"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",[],[]]