[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M78Z7S8GAP4X31FVTQB6VXVG":3,"skill-same:01M78Z7S8GAP4X31FVTQB6VXVG":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},"01M78Z7S8GAP4X31FVTQB6VXVG","evaluation-splitting",null,"Evaluate fitted LightFM recommenders and create leakage-safe train\u002Ftest splits.","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\u002Flightfm\u002Fsub-skills\u002Fevaluation-splitting\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: evaluation-splitting\ndescription: \"Evaluate fitted LightFM recommenders and create leakage-safe\n  train\u002Ftest splits.\"\ndisable-model-invocation: true\nmetadata:\n  disco-role: operating\nlicense: Apache 2.0\n---\n\n# LightFM evaluation and splitting\n\nUse this sub-skill when a task is about offline ranking metrics, train\u002Ftest\nsplits, or debugging leakage in a fitted LightFM recommender.\n\n## Read when\n\n- Compute `precision_at_k`, `recall_at_k`, `auc_score`, `reciprocal_rank`, or\n  `LightFM.predict_rank` for a fitted model.\n- Split an interactions matrix with `random_train_test_split` and validate that\n  train and test interactions do not overlap.\n- Explain `train_interactions`, `check_intersections`, `preserve_rows`, metric\n  output shapes, top-k behavior, rank values, or tie handling.\n- Diagnose intersection `ValueError`, suspiciously high test metrics, empty\n  metric arrays, zero-row users, sparse matrix input problems, or\n  `num_threads` errors.\n\n## Route elsewhere\n\n- Training, tuning, resuming, serializing, direct `predict` scoring, or loss\n  selection: read [model-training](..\u002Fmodel-training\u002FSKILL.md).\n- Dataset mappings, feature matrix construction, built-in dataset fetchers,\n  cold-start feature design, or interaction schemas: read\n  [data-features](..\u002Fdata-features\u002FSKILL.md).\n- Repository build, editable install, compiled extension maintenance, native\n  tests, or CI commands: read [repo-development](..\u002Frepo-development\u002FSKILL.md).\n\n## Fast workflow\n\n1. Start from a fitted `lightfm.LightFM` model and SciPy sparse train\u002Ftest\n   matrices with identical `(n_users, n_items)` shape. If the model was fit\n   with user or item features, pass compatible feature matrices into the metric\n   call as well.\n2. Treat non-zero entries in `test_interactions` as positives to rank. Remove\n   or zero-out negative feedback before ranking metrics; LightFM evaluation\n   treats any non-zero entry as an interaction.\n3. Pass `train_interactions=train` for test metrics so known training positives\n   are excluded from the candidate ranking. Keep `check_intersections=True`\n   unless performing an intentional diagnostic.\n4. Choose `preserve_rows=False` for aggregate means over users with test\n   positives, or `preserve_rows=True` when the output must stay aligned to the\n   original user axis.\n5. Inspect [API details](references\u002Fapi-reference.md) for signatures, return\n   shapes, sparse matrix expectations, intersection checks, and tie semantics.\n6. Use [workflows](references\u002Fworkflows.md) for leakage-safe split\u002Fevaluate\n   recipes, user-preserving metrics, and `predict_rank` interpretation.\n7. Use [troubleshooting](references\u002Ftroubleshooting.md) when metrics error,\n   look optimistic, contain zero\u002Fempty rows, or disagree with expected top-k\n   behavior.\n\nA deterministic, no-network fixture is bundled at\n[scripts\u002Fevaluate_lightfm_fixture.py](scripts\u002Fevaluate_lightfm_fixture.py). It\ntrains a tiny CPU model, computes all ranking metrics, validates a random split,\nand can intentionally demonstrate intersection handling.\n\n## Non-negotiables for leakage-safe evaluation\n\n- Do not average test metrics without checking that train\u002Ftest matrices are\n  disjoint at non-zero coordinates.\n- Do not silence intersection checks just to get a score; fix the split unless\n  the task is explicitly evaluating training-set ranking behavior.\n- Do not compare metrics across runs unless `k`, split policy, feature matrices,\n  `preserve_rows`, and train-interaction filtering are the same.\n- Do not interpret random splits as temporal validation. Use a chronological or\n  user-aware holdout when production evaluation requires it, then still validate\n  intersections before scoring.\n","# LightFM evaluation and splitting\n\nUse this sub-skill when a task is about offline ranking metrics, train\u002Ftest\nsplits, or debugging leakage in a fitted LightFM recommender.\n\n## Read when\n\n- Compute `precision_at_k`, `recall_at_k`, `auc_score`, `reciprocal_rank`, or\n  `LightFM.predict_rank` for a fitted model.\n- Split an interactions matrix with `random_train_test_split` and validate that\n  train and test interactions do not overlap.\n- Explain `train_interactions`, `check_intersections`, `preserve_rows`, metric\n  output shapes, top-k behavior, rank values, or tie handling.\n- Diagnose intersection `ValueError`, suspiciously high test metrics, empty\n  metric arrays, zero-row users, sparse matrix input problems, or\n  `num_threads` errors.\n\n## Route elsewhere\n\n- Training, tuning, resuming, serializing, direct `predict` scoring, or loss\n  selection: read [model-training](..\u002Fmodel-training\u002FSKILL.md).\n- Dataset mappings, feature matrix construction, built-in dataset fetchers,\n  cold-start feature design, or interaction schemas: read\n  [data-features](..\u002Fdata-features\u002FSKILL.md).\n- Repository build, editable install, compiled extension maintenance, native\n  tests, or CI commands: read [repo-development](..\u002Frepo-development\u002FSKILL.md).\n\n## Fast workflow\n\n1. Start from a fitted `lightfm.LightFM` model and SciPy sparse train\u002Ftest\n   matrices with identical `(n_users, n_items)` shape. If the model was fit\n   with user or item features, pass compatible feature matrices into the metric\n   call as well.\n2. Treat non-zero entries in `test_interactions` as positives to rank. Remove\n   or zero-out negative feedback before ranking metrics; LightFM evaluation\n   treats any non-zero entry as an interaction.\n3. Pass `train_interactions=train` for test metrics so known training positives\n   are excluded from the candidate ranking. Keep `check_intersections=True`\n   unless performing an intentional diagnostic.\n4. Choose `preserve_rows=False` for aggregate means over users with test\n   positives, or `preserve_rows=True` when the output must stay aligned to the\n   original user axis.\n5. Inspect [API details](references\u002Fapi-reference.md) for signatures, return\n   shapes, sparse matrix expectations, intersection checks, and tie semantics.\n6. Use [workflows](references\u002Fworkflows.md) for leakage-safe split\u002Fevaluate\n   recipes, user-preserving metrics, and `predict_rank` interpretation.\n7. Use [troubleshooting](references\u002Ftroubleshooting.md) when metrics error,\n   look optimistic, contain zero\u002Fempty rows, or disagree with expected top-k\n   behavior.\n\nA deterministic, no-network fixture is bundled at\n[scripts\u002Fevaluate_lightfm_fixture.py](scripts\u002Fevaluate_lightfm_fixture.py). It\ntrains a tiny CPU model, computes all ranking metrics, validates a random split,\nand can intentionally demonstrate intersection handling.\n\n## Non-negotiables for leakage-safe evaluation\n\n- Do not average test metrics without checking that train\u002Ftest matrices are\n  disjoint at non-zero coordinates.\n- Do not silence intersection checks just to get a score; fix the split unless\n  the task is explicitly evaluating training-set ranking behavior.\n- Do not compare metrics across runs unless `k`, split policy, feature matrices,\n  `preserve_rows`, and train-interaction filtering are the same.\n- Do not interpret random splits as temporal validation. Use a chronological or\n  user-aware holdout when production evaluation requires it, then still validate\n  intersections before scoring.\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\u002Flightfm\u002Fsub-skills\u002Fevaluation-splitting",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 evaluation-splitting",{"tool":40,"command":41},"bunx","bunx skills add VectorSpaceLab\u002FAREX-Skill --skill evaluation-splitting",{"tool":43,"command":44},"pnpm","pnpm dlx skills add VectorSpaceLab\u002FAREX-Skill --skill evaluation-splitting","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Flightfm\u002Fsub-skills\u002Fevaluation-splitting\u002FSKILL.md",[47,53,57,61],{"path":48,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":51,"sourceHtmlUrl":52},"references\u002Fapi-reference.md","REFERENCE","参考文档",8074,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Flightfm\u002Fsub-skills\u002Fevaluation-splitting\u002Freferences\u002Fapi-reference.md",{"path":54,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":55,"sourceHtmlUrl":56},"references\u002Ftroubleshooting.md",8226,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Flightfm\u002Fsub-skills\u002Fevaluation-splitting\u002Freferences\u002Ftroubleshooting.md",{"path":58,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":59,"sourceHtmlUrl":60},"references\u002Fworkflows.md",7722,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Flightfm\u002Fsub-skills\u002Fevaluation-splitting\u002Freferences\u002Fworkflows.md",{"path":62,"resourceType":63,"resourceTypeDesc":64,"sizeBytes":65,"sourceHtmlUrl":66},"scripts\u002Fevaluate_lightfm_fixture.py","SCRIPT","脚本",8707,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Flightfm\u002Fsub-skills\u002Fevaluation-splitting\u002Fscripts\u002Fevaluate_lightfm_fixture.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",[],[]]