[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M792XAWTX3VVT7X0SW3V18SM":3,"skill-same:01M792XAWTX3VVT7X0SW3V18SM":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},"01M792XAWTX3VVT7X0SW3V18SM","sequence-models",null,"Guides pomegranate sequence workflows with MarkovChain, DenseHMM, and SparseHMM, including sequence shapes, variable-length batches, transition setup, Baum-Welch fitting, priors, posterior decoding, and Viterbi.","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\u002Fpomegranate\u002Fsub-skills\u002Fsequence-models\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: sequence-models\ndescription: \"Guides pomegranate sequence workflows with MarkovChain, DenseHMM,\n  and SparseHMM, including sequence shapes, variable-length batches, transition\n  setup, Baum-Welch fitting, priors, posterior decoding, and Viterbi.\"\ndisable-model-invocation: true\nmetadata:\n  disco-role: operating\nlicense: MIT\n---\n\n# sequence-models\n\n## Use this sub-skill when\n\nUse this sub-skill for pomegranate sequence models: k-th order categorical `MarkovChain`, dense-transition `DenseHMM`, sparse-transition `SparseHMM`, sequence likelihood scoring, sequence sampling, hidden-state posterior decoding, Viterbi paths, variable-length sequence training, and prior-weighted\u002Fsemi-supervised HMM learning.\n\n## Start here\n\nTypical imports:\n\n```python\nfrom pomegranate.distributions import Categorical, ConditionalCategorical, Exponential, Normal\nfrom pomegranate.markov_chain import MarkovChain\nfrom pomegranate.hmm import DenseHMM, SparseHMM\n```\n\nRead [references\u002Fapi-reference.md](references\u002Fapi-reference.md) for constructors, sequence data shapes, dense\u002Fsparse edge formats, and examples. Run [scripts\u002Fsmoke_sequence_models.py](scripts\u002Fsmoke_sequence_models.py) to verify tiny MarkovChain, DenseHMM, and SparseHMM workflows.\n\n## Core workflow\n\n1. **Choose the model.** Use `MarkovChain` for observed categorical sequences. Use `DenseHMM` when hidden state transitions are dense. Use `SparseHMM` when only a few state transitions are possible.\n2. **Shape input as sequences.** Most HMM\u002FMarkov-chain scoring uses `(n, length, d)`. HMM fitting can also accept lists of sequences or grouped tensors for variable lengths.\n3. **Create emissions.** HMM states are pomegranate distributions such as `Normal`, `Exponential`, or richer distributions.\n4. **Define starts, ends, and transitions.** Dense HMMs use a transition matrix; sparse HMMs use `[start_distribution, end_distribution, probability]` edge triples plus optional starts\u002Fends.\n5. **Train or score.** Use `fit` for Baum-Welch\u002FEM, `summarize`\u002F`from_summaries` for chunked updates, and `log_probability` for sequence likelihoods.\n6. **Decode states.** Use `predict`, `predict_proba`, `predict_log_proba`, or `viterbi` depending on whether you need posterior marginals or a best path.\n\n## Route elsewhere when\n\n- You need base emissions or categorical distributions: read [..\u002Fdistributions\u002FSKILL.md](..\u002Fdistributions\u002FSKILL.md).\n- You need non-sequential mixtures or classifiers: read [..\u002Fmixtures-and-classifiers\u002FSKILL.md](..\u002Fmixtures-and-classifiers\u002FSKILL.md).\n- You need Bayesian networks or factor graphs over discrete variables: read [..\u002Fgraph-models\u002FSKILL.md](..\u002Fgraph-models\u002FSKILL.md).\n\n## Guardrails\n\n- HMMs are over sequences; do not pass ordinary `(n, d)` arrays to HMM scoring\u002Ffitting unless the method specifically documents that format.\n- `SparseHMM` needs explicit nonzero edges; if every state connects to every state, use `DenseHMM`.\n- Sampling from HMMs requires either fixed `sample_length` or usable end probabilities.\n- Prior tensors for HMMs must align with sequence shape and state count.\n- Read [references\u002Ftroubleshooting.md](references\u002Ftroubleshooting.md) when variable-length data, edge formats, start\u002Fend probabilities, priors, or sampling fail.\n","# sequence-models\n\n## Use this sub-skill when\n\nUse this sub-skill for pomegranate sequence models: k-th order categorical `MarkovChain`, dense-transition `DenseHMM`, sparse-transition `SparseHMM`, sequence likelihood scoring, sequence sampling, hidden-state posterior decoding, Viterbi paths, variable-length sequence training, and prior-weighted\u002Fsemi-supervised HMM learning.\n\n## Start here\n\nTypical imports:\n\n```python\nfrom pomegranate.distributions import Categorical, ConditionalCategorical, Exponential, Normal\nfrom pomegranate.markov_chain import MarkovChain\nfrom pomegranate.hmm import DenseHMM, SparseHMM\n```\n\nRead [references\u002Fapi-reference.md](references\u002Fapi-reference.md) for constructors, sequence data shapes, dense\u002Fsparse edge formats, and examples. Run [scripts\u002Fsmoke_sequence_models.py](scripts\u002Fsmoke_sequence_models.py) to verify tiny MarkovChain, DenseHMM, and SparseHMM workflows.\n\n## Core workflow\n\n1. **Choose the model.** Use `MarkovChain` for observed categorical sequences. Use `DenseHMM` when hidden state transitions are dense. Use `SparseHMM` when only a few state transitions are possible.\n2. **Shape input as sequences.** Most HMM\u002FMarkov-chain scoring uses `(n, length, d)`. HMM fitting can also accept lists of sequences or grouped tensors for variable lengths.\n3. **Create emissions.** HMM states are pomegranate distributions such as `Normal`, `Exponential`, or richer distributions.\n4. **Define starts, ends, and transitions.** Dense HMMs use a transition matrix; sparse HMMs use `[start_distribution, end_distribution, probability]` edge triples plus optional starts\u002Fends.\n5. **Train or score.** Use `fit` for Baum-Welch\u002FEM, `summarize`\u002F`from_summaries` for chunked updates, and `log_probability` for sequence likelihoods.\n6. **Decode states.** Use `predict`, `predict_proba`, `predict_log_proba`, or `viterbi` depending on whether you need posterior marginals or a best path.\n\n## Route elsewhere when\n\n- You need base emissions or categorical distributions: read [..\u002Fdistributions\u002FSKILL.md](..\u002Fdistributions\u002FSKILL.md).\n- You need non-sequential mixtures or classifiers: read [..\u002Fmixtures-and-classifiers\u002FSKILL.md](..\u002Fmixtures-and-classifiers\u002FSKILL.md).\n- You need Bayesian networks or factor graphs over discrete variables: read [..\u002Fgraph-models\u002FSKILL.md](..\u002Fgraph-models\u002FSKILL.md).\n\n## Guardrails\n\n- HMMs are over sequences; do not pass ordinary `(n, d)` arrays to HMM scoring\u002Ffitting unless the method specifically documents that format.\n- `SparseHMM` needs explicit nonzero edges; if every state connects to every state, use `DenseHMM`.\n- Sampling from HMMs requires either fixed `sample_length` or usable end probabilities.\n- Prior tensors for HMMs must align with sequence shape and state count.\n- Read [references\u002Ftroubleshooting.md](references\u002Ftroubleshooting.md) when variable-length data, edge formats, start\u002Fend probabilities, priors, or sampling fail.\n",{"name":5,"license":14,"metadata":23,"description":7,"disable-model-invocation":25},{"disco-role":24},"operating",true,"main","ac3fe1afa80fb9a09775ecfb2b6cc3ba850a2db6","skills\u002Frepositories\u002Frepo-skills\u002Fpomegranate\u002Fsub-skills\u002Fsequence-models",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 sequence-models",{"tool":39,"command":40},"bunx","bunx skills add VectorSpaceLab\u002FAREX-Skill --skill sequence-models",{"tool":42,"command":43},"pnpm","pnpm dlx skills add VectorSpaceLab\u002FAREX-Skill --skill sequence-models","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fpomegranate\u002Fsub-skills\u002Fsequence-models\u002FSKILL.md",[46,52,56],{"path":47,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":50,"sourceHtmlUrl":51},"references\u002Fapi-reference.md","REFERENCE","参考文档",4348,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fpomegranate\u002Fsub-skills\u002Fsequence-models\u002Freferences\u002Fapi-reference.md",{"path":53,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":54,"sourceHtmlUrl":55},"references\u002Ftroubleshooting.md",2939,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fpomegranate\u002Fsub-skills\u002Fsequence-models\u002Freferences\u002Ftroubleshooting.md",{"path":57,"resourceType":58,"resourceTypeDesc":59,"sizeBytes":60,"sourceHtmlUrl":61},"scripts\u002Fsmoke_sequence_models.py","SCRIPT","脚本",1922,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fpomegranate\u002Fsub-skills\u002Fsequence-models\u002Fscripts\u002Fsmoke_sequence_models.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",[],[]]