[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M790KD015RF36YCVVA3RZTNH":3,"skill-same:01M790KD015RF36YCVVA3RZTNH":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},"01M790KD015RF36YCVVA3RZTNH","training-evaluation",null,"Construct and troubleshoot MMDetection3D training, testing, evaluation, visualization-hook, TTA, distributed, and Slurm commands.","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\u002Fmmdetection3d\u002Fsub-skills\u002Ftraining-evaluation\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: training-evaluation\ndescription: \"Construct and troubleshoot MMDetection3D training, testing,\n  evaluation, visualization-hook, TTA, distributed, and Slurm commands.\"\ndisable-model-invocation: true\nmetadata:\n  disco-role: operating\nlicense: Apache 2.0\n---\n\n# MMDetection3D training-evaluation sub-skill\n\nUse this sub-skill when the user needs to train, test, evaluate, visualize evaluation outputs, enable test-time augmentation, or launch distributed\u002FSlurm MMDetection3D jobs from an already chosen config and prepared dataset.\n\n## Best-fit tasks\n\n- Build safe single-process, distributed, or Slurm training commands for `tools\u002Ftrain.py` semantics.\n- Build safe single-process, distributed, or Slurm test\u002Fevaluation commands for `tools\u002Ftest.py` semantics.\n- Decide how to pass `--work-dir`, `--amp`, `--sync_bn`, `--auto-scale-lr`, `--resume`, and nested `--cfg-options` overrides.\n- Add evaluation output prefixes for validation artifacts or benchmark submissions.\n- Add visualization-hook flags (`--show`, `--show-dir`, `--task`, score threshold, wait time) without assuming a GUI is available.\n- Add `--tta` only when the selected segmentation config contains the required TTA sections.\n- Diagnose launch, evaluator, visualization, TTA, CPU\u002FGPU, port, and Slurm failures.\n\n## Route elsewhere\n\n- Dataset download, info-file generation, dataset layout validation, and stale annotation migrations belong in `data-preparation`.\n- Choosing model families, editing config inheritance, changing dataset roots\u002Fclasses, and validating config syntax belong in `configuration-model-zoo`.\n- Standalone inference demos, inferencer APIs, and prediction serialization without evaluation belong in `inference`.\n- Geometry API use and offline interpretation of saved OBJ\u002Fvisualizer artifacts belong in `structures-visualization`.\n- TorchServe, model publishing\u002Fconversion, log plotting, FLOPs, and benchmark helper boundaries belong in `serving-tools`.\n\n## Operating procedure\n\n1. Confirm the caller has a config path, dataset prepared for that config, and for testing a checkpoint path. If these are missing, route before generating commands.\n2. Choose the launcher pattern in [references\u002Fworkflows.md](references\u002Fworkflows.md): single process, distributed shell launcher, or Slurm launcher.\n3. Generate command strings with the bundled dry-run helper:\n\n   ```bash\n   python scripts\u002Fbuild_train_test_command.py --help\n   ```\n\n   The helper prints shell commands only; it never starts training, testing, distributed workers, or Slurm jobs.\n4. For training, decide whether the job needs `--work-dir`, `--amp`, `--sync_bn`, `--auto-scale-lr`, `--resume`, and config overrides before adding launch resources.\n5. For testing\u002Fevaluation, decide whether the task is ordinary metric evaluation, benchmark formatting, saved visualization, or segmentation TTA. Use [references\u002Fevaluation.md](references\u002Fevaluation.md) for evaluator-specific output keys.\n6. For distributed or Slurm jobs, set GPU count and ports explicitly when multiple jobs may share a host. Avoid the default port collision described in [references\u002Ftroubleshooting.md](references\u002Ftroubleshooting.md).\n7. Before recommending an expensive run, check the high-signal guardrails below and the failure patterns in [references\u002Ftroubleshooting.md](references\u002Ftroubleshooting.md).\n\n## High-signal guardrails\n\n- Do not run long training\u002Fevaluation as a verification step. Prefer parser\u002Fhelp checks, config checks, and command rendering unless the user explicitly authorizes the expensive job.\n- `--work-dir` overrides config `work_dir`; if neither is set, MMEngine uses `.\u002Fwork_dirs\u002F\u003Cconfig-basename>`.\n- `--amp` only converts a config whose optimizer wrapper type is `OptimWrapper`; configs already using AMP warn, and unrelated wrappers fail.\n- `--auto-scale-lr` requires the config to define `auto_scale_lr.enable` and `auto_scale_lr.base_batch_size`.\n- `--resume` with no value means auto-resume from the latest checkpoint in the work directory; `--resume path\u002Fto\u002Fcheckpoint.pth` resumes from that checkpoint.\n- `--sync_bn` accepts `none`, `torch`, or `mmcv`; it is mainly useful for multi-GPU training and should not be used as a generic CPU fix.\n- Evaluation output prefixes are usually nested under `test_evaluator.*`; do not pass a top-level `submission_prefix` unless the active config explicitly consumes it.\n- `--show`\u002F`--show-dir` activate the visualization hook and require a valid `--task` choice. Saved visualization works headlessly through output files, but documented use is for single-GPU debugging.\n- `--tta` in this release is for 3D segmentation configs that include both `tta_model` and `tta_pipeline`; do not enable it for ordinary 3D detection configs.\n- CPU train\u002Ftest is experimental and narrow. Most point-cloud models depend on CUDA 3D ops; CPU testing is documented only for SMOKE-style debugging.\n","# MMDetection3D training-evaluation sub-skill\n\nUse this sub-skill when the user needs to train, test, evaluate, visualize evaluation outputs, enable test-time augmentation, or launch distributed\u002FSlurm MMDetection3D jobs from an already chosen config and prepared dataset.\n\n## Best-fit tasks\n\n- Build safe single-process, distributed, or Slurm training commands for `tools\u002Ftrain.py` semantics.\n- Build safe single-process, distributed, or Slurm test\u002Fevaluation commands for `tools\u002Ftest.py` semantics.\n- Decide how to pass `--work-dir`, `--amp`, `--sync_bn`, `--auto-scale-lr`, `--resume`, and nested `--cfg-options` overrides.\n- Add evaluation output prefixes for validation artifacts or benchmark submissions.\n- Add visualization-hook flags (`--show`, `--show-dir`, `--task`, score threshold, wait time) without assuming a GUI is available.\n- Add `--tta` only when the selected segmentation config contains the required TTA sections.\n- Diagnose launch, evaluator, visualization, TTA, CPU\u002FGPU, port, and Slurm failures.\n\n## Route elsewhere\n\n- Dataset download, info-file generation, dataset layout validation, and stale annotation migrations belong in `data-preparation`.\n- Choosing model families, editing config inheritance, changing dataset roots\u002Fclasses, and validating config syntax belong in `configuration-model-zoo`.\n- Standalone inference demos, inferencer APIs, and prediction serialization without evaluation belong in `inference`.\n- Geometry API use and offline interpretation of saved OBJ\u002Fvisualizer artifacts belong in `structures-visualization`.\n- TorchServe, model publishing\u002Fconversion, log plotting, FLOPs, and benchmark helper boundaries belong in `serving-tools`.\n\n## Operating procedure\n\n1. Confirm the caller has a config path, dataset prepared for that config, and for testing a checkpoint path. If these are missing, route before generating commands.\n2. Choose the launcher pattern in [references\u002Fworkflows.md](references\u002Fworkflows.md): single process, distributed shell launcher, or Slurm launcher.\n3. Generate command strings with the bundled dry-run helper:\n\n   ```bash\n   python scripts\u002Fbuild_train_test_command.py --help\n   ```\n\n   The helper prints shell commands only; it never starts training, testing, distributed workers, or Slurm jobs.\n4. For training, decide whether the job needs `--work-dir`, `--amp`, `--sync_bn`, `--auto-scale-lr`, `--resume`, and config overrides before adding launch resources.\n5. For testing\u002Fevaluation, decide whether the task is ordinary metric evaluation, benchmark formatting, saved visualization, or segmentation TTA. Use [references\u002Fevaluation.md](references\u002Fevaluation.md) for evaluator-specific output keys.\n6. For distributed or Slurm jobs, set GPU count and ports explicitly when multiple jobs may share a host. Avoid the default port collision described in [references\u002Ftroubleshooting.md](references\u002Ftroubleshooting.md).\n7. Before recommending an expensive run, check the high-signal guardrails below and the failure patterns in [references\u002Ftroubleshooting.md](references\u002Ftroubleshooting.md).\n\n## High-signal guardrails\n\n- Do not run long training\u002Fevaluation as a verification step. Prefer parser\u002Fhelp checks, config checks, and command rendering unless the user explicitly authorizes the expensive job.\n- `--work-dir` overrides config `work_dir`; if neither is set, MMEngine uses `.\u002Fwork_dirs\u002F\u003Cconfig-basename>`.\n- `--amp` only converts a config whose optimizer wrapper type is `OptimWrapper`; configs already using AMP warn, and unrelated wrappers fail.\n- `--auto-scale-lr` requires the config to define `auto_scale_lr.enable` and `auto_scale_lr.base_batch_size`.\n- `--resume` with no value means auto-resume from the latest checkpoint in the work directory; `--resume path\u002Fto\u002Fcheckpoint.pth` resumes from that checkpoint.\n- `--sync_bn` accepts `none`, `torch`, or `mmcv`; it is mainly useful for multi-GPU training and should not be used as a generic CPU fix.\n- Evaluation output prefixes are usually nested under `test_evaluator.*`; do not pass a top-level `submission_prefix` unless the active config explicitly consumes it.\n- `--show`\u002F`--show-dir` activate the visualization hook and require a valid `--task` choice. Saved visualization works headlessly through output files, but documented use is for single-GPU debugging.\n- `--tta` in this release is for 3D segmentation configs that include both `tta_model` and `tta_pipeline`; do not enable it for ordinary 3D detection configs.\n- CPU train\u002Ftest is experimental and narrow. Most point-cloud models depend on CUDA 3D ops; CPU testing is documented only for SMOKE-style debugging.\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\u002Fmmdetection3d\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":35,"prompt":45,"zipAvailable":26},[36,39,42],{"tool":37,"command":38},"npx","npx skills add VectorSpaceLab\u002FAREX-Skill --skill training-evaluation",{"tool":40,"command":41},"bunx","bunx skills add VectorSpaceLab\u002FAREX-Skill --skill training-evaluation",{"tool":43,"command":44},"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\u002Fmmdetection3d\u002Fsub-skills\u002Ftraining-evaluation\u002FSKILL.md",[47,53,57,61],{"path":48,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":51,"sourceHtmlUrl":52},"references\u002Fevaluation.md","REFERENCE","参考文档",6390,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fmmdetection3d\u002Fsub-skills\u002Ftraining-evaluation\u002Freferences\u002Fevaluation.md",{"path":54,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":55,"sourceHtmlUrl":56},"references\u002Ftroubleshooting.md",6861,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fmmdetection3d\u002Fsub-skills\u002Ftraining-evaluation\u002Freferences\u002Ftroubleshooting.md",{"path":58,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":59,"sourceHtmlUrl":60},"references\u002Fworkflows.md",6493,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fmmdetection3d\u002Fsub-skills\u002Ftraining-evaluation\u002Freferences\u002Fworkflows.md",{"path":62,"resourceType":63,"resourceTypeDesc":64,"sizeBytes":65,"sourceHtmlUrl":66},"scripts\u002Fbuild_train_test_command.py","SCRIPT","脚本",13067,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fmmdetection3d\u002Fsub-skills\u002Ftraining-evaluation\u002Fscripts\u002Fbuild_train_test_command.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",[],[]]