[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M793PX7ZC6HT8TRE2QM033P0":3,"skill-same:01M793PX7ZC6HT8TRE2QM033P0":62},{"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},"01M793PX7ZC6HT8TRE2QM033P0","training",null,"Routes PyTorch Metric Learning questions about trainers, hooks, logging, checkpointing, and optional distributed wrappers.","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\u002Fpytorch-metric-learning\u002Fsub-skills\u002Ftraining\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: training\ndescription: \"Routes PyTorch Metric Learning questions about trainers, hooks,\n  logging, checkpointing, and optional distributed wrappers.\"\ndisable-model-invocation: true\nmetadata:\n  disco-role: operating\nlicense: MIT\n---\n\n# Training\n\nUse this sub-skill when the user wants to wire PyTorch Metric Learning into an actual training loop, validation hook, or checkpoint\u002Flogging workflow.\n\n## Typical triggers\n\n- \"How do I use MetricLossOnly \u002F TrainWithClassifier \u002F CascadedEmbeddings \u002F DeepAdversarialMetricLearning \u002F TwoStreamMetricLoss?\"\n- \"How do I configure the trainer dictionary keys?\"\n- \"How do I add logging, early stopping, or checkpoint saving?\"\n- \"How do I freeze part of the model or use `MPerClassSampler` during training?\"\n- \"How do I wrap a loss or miner for distributed training?\"\n\n## In scope\n\n- Trainers: `BaseTrainer`, `MetricLossOnly`, `TrainWithClassifier`, `CascadedEmbeddings`, `DeepAdversarialMetricLearning`, `TwoStreamMetricLoss`.\n- Hooking and logging: `logging_presets.HookContainer`, record-keeper integration, model saving\u002Floading, early stopping, validation hooks, and metric history access.\n- Trainer configuration: `models`, `optimizers`, `loss_funcs`, `mining_funcs`, `sampler`, `batch_size`, `iterations_per_epoch`, `freeze_these`, `freeze_trunk_batchnorm`, `label_hierarchy_level`, `data_and_label_getter`, `dataset_labels`, `set_min_label_to_zero`, `lr_schedulers`, and `gradient_clippers`.\n- Optional distributed wrappers: `DistributedLossWrapper` and `DistributedMinerWrapper`.\n\n## Out of scope\n\n- Choosing the loss\u002Fminer\u002Fdistance\u002Freducer\u002Fregularizer stack belongs in `components`.\n- Accuracy calculators, testers, and nearest-neighbor search belong in `evaluation`.\n- Downloading datasets or building samplers belongs in `data` unless the question is purely about their role inside a trainer.\n\n## How to use this sub-skill\n\n1. Read `references\u002Ftraining-workflows.md` for the trainer map, hook sequence, and config expectations.\n2. Run `scripts\u002Fsmoke_training.py` when you want to confirm the trainer wiring on toy data without downloads.\n3. Read `references\u002Ftroubleshooting.md` when the failure mentions missing dictionary keys, frozen modules, checkpointing, or logging dependencies.\n4. If the request mentions distributed loss\u002Fminer wrappers, read the optional distributed note in the workflow reference and treat GPU execution as optional unless the user specifically asks for it.\n\n## Common routing decisions\n\n- If the user only needs the right loss or miner, route to `components` first.\n- If the user wants retrieval metrics, validation scores, or nearest-neighbor search after training, route to `evaluation` after the trainer is set up.\n- If the user wants dataset download or batch construction help, route to `data` before wiring the trainer.\n\n## Useful public facts\n\n- All trainer subclasses share the `BaseTrainer` argument pattern.\n- `MetricLossOnly` needs a trunk model and a `metric_loss` entry.\n- `TrainWithClassifier` expects a trunk\u002Fclassifier architecture and both metric and classification losses.\n- `CascadedEmbeddings` splits the embedding dimension into multiple stages.\n- `DeepAdversarialMetricLearning` adds generator-specific loss keys and epoch scheduling.\n- `TwoStreamMetricLoss` expects a dataset that yields `(anchor, positive, label)`.\n- `HookContainer` is the easiest path for record-keeper, tensorboard, early stopping, and model checkpointing.\n\n## Read next\n\n- `references\u002Ftraining-workflows.md` for trainer and hook details.\n- `references\u002Ftroubleshooting.md` for trainer-key, checkpoint, logging, and distributed-wrapper failures.\n- `scripts\u002Fsmoke_training.py` for a tiny end-to-end trainer smoke check.\n","# Training\n\nUse this sub-skill when the user wants to wire PyTorch Metric Learning into an actual training loop, validation hook, or checkpoint\u002Flogging workflow.\n\n## Typical triggers\n\n- \"How do I use MetricLossOnly \u002F TrainWithClassifier \u002F CascadedEmbeddings \u002F DeepAdversarialMetricLearning \u002F TwoStreamMetricLoss?\"\n- \"How do I configure the trainer dictionary keys?\"\n- \"How do I add logging, early stopping, or checkpoint saving?\"\n- \"How do I freeze part of the model or use `MPerClassSampler` during training?\"\n- \"How do I wrap a loss or miner for distributed training?\"\n\n## In scope\n\n- Trainers: `BaseTrainer`, `MetricLossOnly`, `TrainWithClassifier`, `CascadedEmbeddings`, `DeepAdversarialMetricLearning`, `TwoStreamMetricLoss`.\n- Hooking and logging: `logging_presets.HookContainer`, record-keeper integration, model saving\u002Floading, early stopping, validation hooks, and metric history access.\n- Trainer configuration: `models`, `optimizers`, `loss_funcs`, `mining_funcs`, `sampler`, `batch_size`, `iterations_per_epoch`, `freeze_these`, `freeze_trunk_batchnorm`, `label_hierarchy_level`, `data_and_label_getter`, `dataset_labels`, `set_min_label_to_zero`, `lr_schedulers`, and `gradient_clippers`.\n- Optional distributed wrappers: `DistributedLossWrapper` and `DistributedMinerWrapper`.\n\n## Out of scope\n\n- Choosing the loss\u002Fminer\u002Fdistance\u002Freducer\u002Fregularizer stack belongs in `components`.\n- Accuracy calculators, testers, and nearest-neighbor search belong in `evaluation`.\n- Downloading datasets or building samplers belongs in `data` unless the question is purely about their role inside a trainer.\n\n## How to use this sub-skill\n\n1. Read `references\u002Ftraining-workflows.md` for the trainer map, hook sequence, and config expectations.\n2. Run `scripts\u002Fsmoke_training.py` when you want to confirm the trainer wiring on toy data without downloads.\n3. Read `references\u002Ftroubleshooting.md` when the failure mentions missing dictionary keys, frozen modules, checkpointing, or logging dependencies.\n4. If the request mentions distributed loss\u002Fminer wrappers, read the optional distributed note in the workflow reference and treat GPU execution as optional unless the user specifically asks for it.\n\n## Common routing decisions\n\n- If the user only needs the right loss or miner, route to `components` first.\n- If the user wants retrieval metrics, validation scores, or nearest-neighbor search after training, route to `evaluation` after the trainer is set up.\n- If the user wants dataset download or batch construction help, route to `data` before wiring the trainer.\n\n## Useful public facts\n\n- All trainer subclasses share the `BaseTrainer` argument pattern.\n- `MetricLossOnly` needs a trunk model and a `metric_loss` entry.\n- `TrainWithClassifier` expects a trunk\u002Fclassifier architecture and both metric and classification losses.\n- `CascadedEmbeddings` splits the embedding dimension into multiple stages.\n- `DeepAdversarialMetricLearning` adds generator-specific loss keys and epoch scheduling.\n- `TwoStreamMetricLoss` expects a dataset that yields `(anchor, positive, label)`.\n- `HookContainer` is the easiest path for record-keeper, tensorboard, early stopping, and model checkpointing.\n\n## Read next\n\n- `references\u002Ftraining-workflows.md` for trainer and hook details.\n- `references\u002Ftroubleshooting.md` for trainer-key, checkpoint, logging, and distributed-wrapper failures.\n- `scripts\u002Fsmoke_training.py` for a tiny end-to-end trainer smoke check.\n",{"name":5,"license":14,"metadata":23,"description":7,"disable-model-invocation":25},{"disco-role":24},"operating",true,"main","ac3fe1afa80fb9a09775ecfb2b6cc3ba850a2db6","skills\u002Frepositories\u002Frepo-skills\u002Fpytorch-metric-learning\u002Fsub-skills\u002Ftraining",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 training",{"tool":39,"command":40},"bunx","bunx skills add VectorSpaceLab\u002FAREX-Skill --skill training",{"tool":42,"command":43},"pnpm","pnpm dlx skills add VectorSpaceLab\u002FAREX-Skill --skill training","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fpytorch-metric-learning\u002Fsub-skills\u002Ftraining\u002FSKILL.md",[46,52,56],{"path":47,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":50,"sourceHtmlUrl":51},"references\u002Ftraining-workflows.md","REFERENCE","参考文档",4758,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fpytorch-metric-learning\u002Fsub-skills\u002Ftraining\u002Freferences\u002Ftraining-workflows.md",{"path":53,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":54,"sourceHtmlUrl":55},"references\u002Ftroubleshooting.md",2233,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fpytorch-metric-learning\u002Fsub-skills\u002Ftraining\u002Freferences\u002Ftroubleshooting.md",{"path":57,"resourceType":58,"resourceTypeDesc":59,"sizeBytes":60,"sourceHtmlUrl":61},"scripts\u002Fsmoke_training.py","SCRIPT","脚本",3265,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fpytorch-metric-learning\u002Fsub-skills\u002Ftraining\u002Fscripts\u002Fsmoke_training.py",[63,71,78,85,92,99],{"skillNo":64,"name":65,"nameZh":6,"description":66,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":67,"licenseFileUrl":6,"sourceHtmlUrl":68,"categories":69,"keywords":70,"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":72,"name":73,"nameZh":6,"description":74,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":67,"licenseFileUrl":6,"sourceHtmlUrl":75,"categories":76,"keywords":77,"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":79,"name":80,"nameZh":6,"description":81,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":67,"licenseFileUrl":6,"sourceHtmlUrl":82,"categories":83,"keywords":84,"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":86,"name":87,"nameZh":6,"description":88,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":67,"licenseFileUrl":6,"sourceHtmlUrl":89,"categories":90,"keywords":91,"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":93,"name":94,"nameZh":6,"description":95,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":67,"licenseFileUrl":6,"sourceHtmlUrl":96,"categories":97,"keywords":98,"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":100,"name":101,"nameZh":6,"description":102,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":67,"licenseFileUrl":6,"sourceHtmlUrl":103,"categories":104,"keywords":105,"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",[],[]]