[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M796F5YK48RV1NHTZ604C9SZ":3,"skill-same:01M796F5YK48RV1NHTZ604C9SZ":74},{"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},"01M796F5YK48RV1NHTZ604C9SZ","objectives-and-training",null,"Choose and wire TorchRL losses, value estimators, target updaters, trainers, Hydra configs, and SOTA recipes.","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\u002Ftorchrl\u002Fsub-skills\u002Fobjectives-and-training\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: objectives-and-training\ndescription: \"Choose and wire TorchRL losses, value estimators, target updaters,\n  trainers, Hydra configs, and SOTA recipes.\"\ndisable-model-invocation: true\nmetadata:\n  disco-role: operating\nlicense: MIT\n---\n\n# TorchRL objectives and training\n\nUse this sub-skill when a task is about TorchRL objective modules, value estimators, target-network updates, trainer hooks, Hydra algorithm configs, or choosing an algorithm recipe. Typical triggers: `ClipPPOLoss`, `PPOLoss`, `SACLoss`, `DQNLoss`, `DDPGLoss`, `TD3Loss`, `IQLLoss`, `CQLLoss`, `MAPPOLoss`, `ValueEstimators`, `make_value_estimator`, `set_keys`, `_AcceptedKeys`, `SoftUpdate`, `HardUpdate`, `Trainer`, trainer hooks, config parity, and `sota-implementations`.\n\n## Route before acting\n\n- Actor, critic, distribution, recurrent-module, and policy-wrapper construction belongs to sibling sub-skill `modules-and-policies`. Return here after the modules expose the TensorDict keys the loss expects.\n- Collector topology, replay buffers, prioritized samplers, and data movement belong to sibling sub-skill `collectors-and-replay`. Return here for loss-side priority keys and target updates.\n- Environment specs, action specs, `step_mdp`, simulator setup, and transforms belong to sibling sub-skill `envs-and-transforms`.\n- LLM GRPO\u002FSFT\u002FRLHF and VLA-specific serving or dataset workflows belong to sibling sub-skill `llm-vla-and-services`; this sub-skill only covers generic loss-key and trainer conventions that those workflows share.\n\n## Operating sequence\n\n1. Pick the loss family from [loss API reference](references\u002Floss-api-reference.md) by algorithm, action space, online\u002Foffline setting, and multi-agent layout.\n2. Verify the actor\u002Fcritic\u002Freplay\u002Fenv dependencies before writing training code: module `out_keys`, environment `action_spec`, replay sample keys, and `(\"next\", ...)` transition fields must match the loss key map.\n3. Configure TensorDict keys with `loss.set_keys(...)` using the accepted key names listed in [loss API reference](references\u002Floss-api-reference.md). Prefer `NestedKey` tuples for nested fields.\n4. Build the value estimator with `loss.make_value_estimator(ValueEstimators.\u003Cname>, ...)` or a standalone estimator such as `GAE`; make sure reward\u002Fdone\u002Fterminated keys are forwarded to the estimator.\n5. Add target-network updaters when the loss has delayed target parameters, and call updater `step()` at the correct cadence; see [training algorithms](references\u002Ftraining-algorithms.md).\n6. If using TorchRL `Trainer`, register hooks deliberately and follow config\u002Fclass parity rules from [trainer and configs](references\u002Ftrainer-and-configs.md).\n7. Use [SOTA implementation map](references\u002Fsota-implementation-map.md) as a reference-only recipe index. Do not run long SOTA launchers as generic helper scripts.\n8. For failures, start with [troubleshooting](references\u002Ftroubleshooting.md), then use the bundled inspection helper.\n\n## Bundled helper\n\n- [scripts\u002Finspect_loss_keys.py](scripts\u002Finspect_loss_keys.py) prints a named loss class signature, whether `set_keys` is available, default configurable TensorDict keys when accessible, and the default value estimator. It performs no training and is safe to run from any current directory.\n\n## Verification boundary\n\nThe skill content is based on repository source, public reference docs, tutorials, SOTA recipe files, objective tests, trainer\u002Fconfig code, and CPU package\u002FAPI inspection. Core objective\u002Ftrainer APIs are CPU-verifiable. CUDA kernels, Triton paths, distributed learner backends, simulator-heavy examples, and long SOTA training are optional or reference-only unless a future task provisions those dependencies explicitly.\n","# TorchRL objectives and training\n\nUse this sub-skill when a task is about TorchRL objective modules, value estimators, target-network updates, trainer hooks, Hydra algorithm configs, or choosing an algorithm recipe. Typical triggers: `ClipPPOLoss`, `PPOLoss`, `SACLoss`, `DQNLoss`, `DDPGLoss`, `TD3Loss`, `IQLLoss`, `CQLLoss`, `MAPPOLoss`, `ValueEstimators`, `make_value_estimator`, `set_keys`, `_AcceptedKeys`, `SoftUpdate`, `HardUpdate`, `Trainer`, trainer hooks, config parity, and `sota-implementations`.\n\n## Route before acting\n\n- Actor, critic, distribution, recurrent-module, and policy-wrapper construction belongs to sibling sub-skill `modules-and-policies`. Return here after the modules expose the TensorDict keys the loss expects.\n- Collector topology, replay buffers, prioritized samplers, and data movement belong to sibling sub-skill `collectors-and-replay`. Return here for loss-side priority keys and target updates.\n- Environment specs, action specs, `step_mdp`, simulator setup, and transforms belong to sibling sub-skill `envs-and-transforms`.\n- LLM GRPO\u002FSFT\u002FRLHF and VLA-specific serving or dataset workflows belong to sibling sub-skill `llm-vla-and-services`; this sub-skill only covers generic loss-key and trainer conventions that those workflows share.\n\n## Operating sequence\n\n1. Pick the loss family from [loss API reference](references\u002Floss-api-reference.md) by algorithm, action space, online\u002Foffline setting, and multi-agent layout.\n2. Verify the actor\u002Fcritic\u002Freplay\u002Fenv dependencies before writing training code: module `out_keys`, environment `action_spec`, replay sample keys, and `(\"next\", ...)` transition fields must match the loss key map.\n3. Configure TensorDict keys with `loss.set_keys(...)` using the accepted key names listed in [loss API reference](references\u002Floss-api-reference.md). Prefer `NestedKey` tuples for nested fields.\n4. Build the value estimator with `loss.make_value_estimator(ValueEstimators.\u003Cname>, ...)` or a standalone estimator such as `GAE`; make sure reward\u002Fdone\u002Fterminated keys are forwarded to the estimator.\n5. Add target-network updaters when the loss has delayed target parameters, and call updater `step()` at the correct cadence; see [training algorithms](references\u002Ftraining-algorithms.md).\n6. If using TorchRL `Trainer`, register hooks deliberately and follow config\u002Fclass parity rules from [trainer and configs](references\u002Ftrainer-and-configs.md).\n7. Use [SOTA implementation map](references\u002Fsota-implementation-map.md) as a reference-only recipe index. Do not run long SOTA launchers as generic helper scripts.\n8. For failures, start with [troubleshooting](references\u002Ftroubleshooting.md), then use the bundled inspection helper.\n\n## Bundled helper\n\n- [scripts\u002Finspect_loss_keys.py](scripts\u002Finspect_loss_keys.py) prints a named loss class signature, whether `set_keys` is available, default configurable TensorDict keys when accessible, and the default value estimator. It performs no training and is safe to run from any current directory.\n\n## Verification boundary\n\nThe skill content is based on repository source, public reference docs, tutorials, SOTA recipe files, objective tests, trainer\u002Fconfig code, and CPU package\u002FAPI inspection. Core objective\u002Ftrainer APIs are CPU-verifiable. CUDA kernels, Triton paths, distributed learner backends, simulator-heavy examples, and long SOTA training are optional or reference-only unless a future task provisions those dependencies explicitly.\n",{"name":5,"license":14,"metadata":23,"description":7,"disable-model-invocation":25},{"disco-role":24},"operating",true,"main","ac3fe1afa80fb9a09775ecfb2b6cc3ba850a2db6","skills\u002Frepositories\u002Frepo-skills\u002Ftorchrl\u002Fsub-skills\u002Fobjectives-and-training",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 objectives-and-training",{"tool":39,"command":40},"bunx","bunx skills add VectorSpaceLab\u002FAREX-Skill --skill objectives-and-training",{"tool":42,"command":43},"pnpm","pnpm dlx skills add VectorSpaceLab\u002FAREX-Skill --skill objectives-and-training","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Ftorchrl\u002Fsub-skills\u002Fobjectives-and-training\u002FSKILL.md",[46,52,56,60,64,68],{"path":47,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":50,"sourceHtmlUrl":51},"references\u002Floss-api-reference.md","REFERENCE","参考文档",13444,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Ftorchrl\u002Fsub-skills\u002Fobjectives-and-training\u002Freferences\u002Floss-api-reference.md",{"path":53,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":54,"sourceHtmlUrl":55},"references\u002Fsota-implementation-map.md",9637,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Ftorchrl\u002Fsub-skills\u002Fobjectives-and-training\u002Freferences\u002Fsota-implementation-map.md",{"path":57,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":58,"sourceHtmlUrl":59},"references\u002Ftrainer-and-configs.md",8927,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Ftorchrl\u002Fsub-skills\u002Fobjectives-and-training\u002Freferences\u002Ftrainer-and-configs.md",{"path":61,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":62,"sourceHtmlUrl":63},"references\u002Ftraining-algorithms.md",9515,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Ftorchrl\u002Fsub-skills\u002Fobjectives-and-training\u002Freferences\u002Ftraining-algorithms.md",{"path":65,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":66,"sourceHtmlUrl":67},"references\u002Ftroubleshooting.md",9718,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Ftorchrl\u002Fsub-skills\u002Fobjectives-and-training\u002Freferences\u002Ftroubleshooting.md",{"path":69,"resourceType":70,"resourceTypeDesc":71,"sizeBytes":72,"sourceHtmlUrl":73},"scripts\u002Finspect_loss_keys.py","SCRIPT","脚本",6304,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Ftorchrl\u002Fsub-skills\u002Fobjectives-and-training\u002Fscripts\u002Finspect_loss_keys.py",[75,83,90,97,104,111],{"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":79,"licenseFileUrl":6,"sourceHtmlUrl":80,"categories":81,"keywords":82,"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":84,"name":85,"nameZh":6,"description":86,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":79,"licenseFileUrl":6,"sourceHtmlUrl":87,"categories":88,"keywords":89,"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":91,"name":92,"nameZh":6,"description":93,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":79,"licenseFileUrl":6,"sourceHtmlUrl":94,"categories":95,"keywords":96,"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":98,"name":99,"nameZh":6,"description":100,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":79,"licenseFileUrl":6,"sourceHtmlUrl":101,"categories":102,"keywords":103,"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":105,"name":106,"nameZh":6,"description":107,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":79,"licenseFileUrl":6,"sourceHtmlUrl":108,"categories":109,"keywords":110,"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":112,"name":113,"nameZh":6,"description":114,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":79,"licenseFileUrl":6,"sourceHtmlUrl":115,"categories":116,"keywords":117,"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",[],[]]