[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M79476VM2S8T52RSPFKR0Z9G":3,"skill-same:01M79476VM2S8T52RSPFKR0Z9G":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},"01M79476VM2S8T52RSPFKR0Z9G","learning",null,"Configure REINVENT4 transfer learning and staged reinforcement learning runs safely.","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\u002Freinvent4\u002Fsub-skills\u002Flearning\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: learning\ndescription: \"Configure REINVENT4 transfer learning and staged reinforcement\n  learning runs safely.\"\ndisable-model-invocation: true\nmetadata:\n  disco-role: operating\nlicense: Apache 2.0\n---\n\n# REINVENT4 Learning\n\nUse this sub-skill when the user asks for transfer learning, `transfer_learning.toml`, staged RL\u002Fcurriculum learning, `staged_learning.toml`, diversity filters, inception memory, TensorBoard monitoring, or a TL-then-RL workflow.\n\n## Route the request\n\n- Use `run_type = \"transfer_learning\"` to bias a prior or checkpoint toward a curated SMILES set and write an `output_model_file` that can later be sampled or used as an RL `agent_file`.\n- Use `run_type = \"staged_learning\"` for all reinforcement learning, including a single-stage RL run and multi-stage curriculum learning.\n- For TL followed by RL, prepare and validate the TL config first, plan a brief sample\u002Fquality check of the TL model, then set the staged-learning `agent_file` to the chosen TL `.model` or checkpoint while keeping `prior_file` compatible with the same generator type.\n- Cross-link to the scoring sub-skill for scoring component design, transforms, external scoring files, and scoring-only validation; this sub-skill only covers how scoring is embedded in staged learning.\n- Cross-link to the sampling sub-skill when the user needs to sample from a prior, TL model, or RL checkpoint.\n- Cross-link to the data-pipeline sub-skill when the training or seed SMILES file needs cleaning, splitting, column extraction, or standardization.\n\n## Safe workflow\n\n1. Identify the generator family from the model file and input shape: Reinvent uses one-column SMILES; Mol2Mol\u002FPepinvent usually need conditional input plus `sample_strategy`; LibInvent\u002FLinkInvent use two-column fragment inputs.\n2. Use CPU by default for dry-run validation and small examples; if the config says `device = \"cuda:0\"`, tell the user `reinvent --device cpu FILE` overrides the config for CPU fallback.\n3. Validate config shape before any training: run `python sub-skills\u002Flearning\u002Fscripts\u002Fcheck_learning_config.py CONFIG` from the skill root or copy the script beside the planned config.\n4. Never start long TL\u002FRL training without explicit user approval for runtime, model files, input SMILES, scoring files, output locations, and device.\n5. Prefer short exploratory settings first: small `num_epochs` for TL, conservative RL `max_steps`, and checkpoint files on every stage.\n\n## Required config anchors\n\n- TL config: top-level `run_type`, `device`, optional top-level `tb_logdir`, then `[parameters]` with `input_model_file`, `smiles_file`, `output_model_file`, `num_epochs`, `batch_size`, and usually `validation_smiles_file` plus `save_every_n_epochs`.\n- Mol2Mol TL: include `[parameters.pairs]` or dotted `pairs.*` thresholds for Tanimoto pair construction.\n- RL config: top-level `run_type`, `device`, optional top-level `tb_logdir`, `[parameters]`, `[learning_strategy]`, optional `[diversity_filter]`, optional `[inception]` or `[intrinsic_penalty]`, and one or more `[[stage]]` blocks.\n- RL parameters: `prior_file`, `agent_file`, `summary_csv_prefix`, `batch_size`, optional `smiles_file` for conditional generators, and safe checkpoint\u002Fcontinuation settings (`use_checkpoint`, `purge_memories`).\n- RL stages: each `[[stage]]` needs `max_steps`, `termination = \"simple\"`, `min_steps`, `max_score`, optional `chkpt_file`, and `[stage.scoring]` with inline components or `filename` plus `filetype` for bundled stage scoring files.\n\n## Bundled guidance\n\n- Read `references\u002Ftl-rl-workflows.md` for concrete TL, staged RL, diversity filter, inception, intrinsic penalty, and TL-then-RL patterns.\n- Read `references\u002Fmonitoring-and-outputs.md` for TensorBoard log layout, CSV columns, checkpoint behavior, and review criteria.\n- Read `references\u002Ftroubleshooting.md` for device fallback, missing model\u002Finput files, model type mismatch, scoring-file path errors, SMILES column issues, TensorBoard confusion, remote responder setup, and long-run safeguards.\n- Run `scripts\u002Fcheck_learning_config.py` to statically parse TOML\u002FJSON\u002FYAML configs and report missing major sections or risky settings without importing REINVENT4 or starting training.\n","# REINVENT4 Learning\n\nUse this sub-skill when the user asks for transfer learning, `transfer_learning.toml`, staged RL\u002Fcurriculum learning, `staged_learning.toml`, diversity filters, inception memory, TensorBoard monitoring, or a TL-then-RL workflow.\n\n## Route the request\n\n- Use `run_type = \"transfer_learning\"` to bias a prior or checkpoint toward a curated SMILES set and write an `output_model_file` that can later be sampled or used as an RL `agent_file`.\n- Use `run_type = \"staged_learning\"` for all reinforcement learning, including a single-stage RL run and multi-stage curriculum learning.\n- For TL followed by RL, prepare and validate the TL config first, plan a brief sample\u002Fquality check of the TL model, then set the staged-learning `agent_file` to the chosen TL `.model` or checkpoint while keeping `prior_file` compatible with the same generator type.\n- Cross-link to the scoring sub-skill for scoring component design, transforms, external scoring files, and scoring-only validation; this sub-skill only covers how scoring is embedded in staged learning.\n- Cross-link to the sampling sub-skill when the user needs to sample from a prior, TL model, or RL checkpoint.\n- Cross-link to the data-pipeline sub-skill when the training or seed SMILES file needs cleaning, splitting, column extraction, or standardization.\n\n## Safe workflow\n\n1. Identify the generator family from the model file and input shape: Reinvent uses one-column SMILES; Mol2Mol\u002FPepinvent usually need conditional input plus `sample_strategy`; LibInvent\u002FLinkInvent use two-column fragment inputs.\n2. Use CPU by default for dry-run validation and small examples; if the config says `device = \"cuda:0\"`, tell the user `reinvent --device cpu FILE` overrides the config for CPU fallback.\n3. Validate config shape before any training: run `python sub-skills\u002Flearning\u002Fscripts\u002Fcheck_learning_config.py CONFIG` from the skill root or copy the script beside the planned config.\n4. Never start long TL\u002FRL training without explicit user approval for runtime, model files, input SMILES, scoring files, output locations, and device.\n5. Prefer short exploratory settings first: small `num_epochs` for TL, conservative RL `max_steps`, and checkpoint files on every stage.\n\n## Required config anchors\n\n- TL config: top-level `run_type`, `device`, optional top-level `tb_logdir`, then `[parameters]` with `input_model_file`, `smiles_file`, `output_model_file`, `num_epochs`, `batch_size`, and usually `validation_smiles_file` plus `save_every_n_epochs`.\n- Mol2Mol TL: include `[parameters.pairs]` or dotted `pairs.*` thresholds for Tanimoto pair construction.\n- RL config: top-level `run_type`, `device`, optional top-level `tb_logdir`, `[parameters]`, `[learning_strategy]`, optional `[diversity_filter]`, optional `[inception]` or `[intrinsic_penalty]`, and one or more `[[stage]]` blocks.\n- RL parameters: `prior_file`, `agent_file`, `summary_csv_prefix`, `batch_size`, optional `smiles_file` for conditional generators, and safe checkpoint\u002Fcontinuation settings (`use_checkpoint`, `purge_memories`).\n- RL stages: each `[[stage]]` needs `max_steps`, `termination = \"simple\"`, `min_steps`, `max_score`, optional `chkpt_file`, and `[stage.scoring]` with inline components or `filename` plus `filetype` for bundled stage scoring files.\n\n## Bundled guidance\n\n- Read `references\u002Ftl-rl-workflows.md` for concrete TL, staged RL, diversity filter, inception, intrinsic penalty, and TL-then-RL patterns.\n- Read `references\u002Fmonitoring-and-outputs.md` for TensorBoard log layout, CSV columns, checkpoint behavior, and review criteria.\n- Read `references\u002Ftroubleshooting.md` for device fallback, missing model\u002Finput files, model type mismatch, scoring-file path errors, SMILES column issues, TensorBoard confusion, remote responder setup, and long-run safeguards.\n- Run `scripts\u002Fcheck_learning_config.py` to statically parse TOML\u002FJSON\u002FYAML configs and report missing major sections or risky settings without importing REINVENT4 or starting training.\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\u002Freinvent4\u002Fsub-skills\u002Flearning",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 learning",{"tool":40,"command":41},"bunx","bunx skills add VectorSpaceLab\u002FAREX-Skill --skill learning",{"tool":43,"command":44},"pnpm","pnpm dlx skills add VectorSpaceLab\u002FAREX-Skill --skill learning","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Freinvent4\u002Fsub-skills\u002Flearning\u002FSKILL.md",[47,53,57,61],{"path":48,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":51,"sourceHtmlUrl":52},"references\u002Fmonitoring-and-outputs.md","REFERENCE","参考文档",5717,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Freinvent4\u002Fsub-skills\u002Flearning\u002Freferences\u002Fmonitoring-and-outputs.md",{"path":54,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":55,"sourceHtmlUrl":56},"references\u002Ftl-rl-workflows.md",8223,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Freinvent4\u002Fsub-skills\u002Flearning\u002Freferences\u002Ftl-rl-workflows.md",{"path":58,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":59,"sourceHtmlUrl":60},"references\u002Ftroubleshooting.md",7179,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Freinvent4\u002Fsub-skills\u002Flearning\u002Freferences\u002Ftroubleshooting.md",{"path":62,"resourceType":63,"resourceTypeDesc":64,"sizeBytes":65,"sourceHtmlUrl":66},"scripts\u002Fcheck_learning_config.py","SCRIPT","脚本",15058,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Freinvent4\u002Fsub-skills\u002Flearning\u002Fscripts\u002Fcheck_learning_config.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",[],[]]