[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M797KB76FHBFR88CG1WFC303":3,"skill-same:01M797KB76FHBFR88CG1WFC303":71},{"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},"01M797KB76FHBFR88CG1WFC303","text-and-datasets",null,"Use Yellowbrick dataset loaders, cache controls, and text visualizers for tabular data, corpora, and text embeddings.","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\u002Fyellowbrick\u002Fsub-skills\u002Ftext-and-datasets\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: text-and-datasets\ndescription: \"Use Yellowbrick dataset loaders, cache controls, and text\n  visualizers for tabular data, corpora, and text embeddings.\"\ndisable-model-invocation: true\nmetadata:\n  disco-role: operating\nlicense: Apache 2.0\n---\n\n# Yellowbrick Text and Datasets\n\nUse this sub-skill when the task is about Yellowbrick dataset loading, dataset\ncache\u002Fdownload behavior, the `load_*` helpers, the hobbies corpus, or\ntext-specific visualizers. It covers `FreqDistVisualizer`, `TSNEVisualizer`,\n`UMAPVisualizer`, `DispersionPlot`, `WordCorrelationPlot`, and\n`PosTagVisualizer`.\n\nFor package-wide Yellowbrick lifecycle, axes reuse, style, and headless saving,\nread [shared visualizer patterns](..\u002F..\u002Freferences\u002Fvisualizer-patterns.md). For\nbroad install, Matplotlib backend\u002Ffont, and scikit-learn compatibility failures,\nread the root [troubleshooting reference](..\u002F..\u002Freferences\u002Ftroubleshooting.md)\nbefore applying text\u002Fdataset-specific fixes.\n\n## Use this sub-skill for\n\n- Choosing among Yellowbrick example datasets and matching them to regression,\n  classification, clustering, time-series-style, or text-analysis tasks.\n- Loading tabular datasets with `load_bikeshare()`, `load_concrete()`,\n  `load_credit()`, `load_energy()`, `load_game()`, `load_mushroom()`,\n  `load_occupancy()`, `load_spam()`, `load_walking()`, or `load_nfl()`.\n- Loading the hobbies corpus with `load_hobbies()` and using `Corpus.data`,\n  `Corpus.target`, `Corpus.labels`, or `Corpus.files`.\n- Deciding between default `X, y` returns, `return_dataset=True`, `to_numpy()`,\n  `to_pandas()`, and `to_dataframe()`.\n- Controlling or inspecting the dataset cache with `data_home`,\n  `YELLOWBRICK_DATA`, and the no-download cache inspector.\n- Using text visualizers after the corpus is already vectorized, tokenized, or\n  part-of-speech tagged.\n- Explaining optional extras: `pandas` for DataFrame returns, `umap-learn` for\n  `UMAPVisualizer`, and `nltk`\u002F`spacy` plus data\u002Fmodels for parser-based POS\n  tagging.\n\n## Route elsewhere when\n\n- The user has already loaded data and now wants classifier score plots such as\n  reports, confusion matrices, ROC\u002FPR curves, or threshold tuning: route to\n  [classifier visualizers](..\u002Fclassifier-visualizers\u002FSKILL.md).\n- The user wants residuals, prediction error, Cook's distance, or alpha\n  selection: route to [regressor visualizers](..\u002Fregressor-visualizers\u002FSKILL.md).\n- The user wants feature ranking, PCA\u002Fmanifold feature projections, class\n  balance, binning, or feature-target correlation: route to\n  [feature-target visualizers](..\u002Ffeature-target-visualizers\u002FSKILL.md).\n- The user wants clustering diagnostics, validation\u002Flearning curves, CV scores,\n  RFECV, feature importances, or dropping curves: route to\n  [cluster-model-selection](..\u002Fcluster-model-selection\u002FSKILL.md).\n- The user asks about contrib wrappers, decision boundaries, missing-value\n  contrib plots, or statsmodels adapters: route to\n  [contrib-and-extensions](..\u002Fcontrib-and-extensions\u002FSKILL.md).\n\n## Required read and run map\n\n| Need | Read or run | When to use it |\n|---|---|---|\n| Dataset names, task fit, `load_*`, return types, `Dataset`\u002F`Corpus`, cache rules, downloader flags, and offline behavior | [datasets reference](references\u002Fdatasets.md) | Read before giving loader code, cache advice, or no-network guidance. |\n| Text visualizer selection, exact input formats, optional extras, and safe file output | [text visualizers reference](references\u002Ftext-visualizers.md) | Read before using any `yellowbrick.text` visualizer or diagnosing text input-shape errors. |\n| Dataset cache\u002Fdownload\u002Fsignature\u002Foffline failures, missing optional dependencies, wrong class names, parser\u002Ftagset errors, and display\u002Ffont warnings | [text\u002Fdataset troubleshooting](references\u002Ftroubleshooting.md) | Read after any failure symptom or before advising a risky cleanup\u002Fdownload action. |\n| Common Yellowbrick `fit`\u002F`score`\u002F`transform`\u002F`show(outpath=...)`, Matplotlib `Agg`, axes, and style rules | [shared visualizer patterns](..\u002F..\u002Freferences\u002Fvisualizer-patterns.md) | Read before writing report-quality code or headless automation. |\n| Broad installation, backend, font, or scikit-learn compatibility issues | [root troubleshooting](..\u002F..\u002Freferences\u002Ftroubleshooting.md) | Read when the problem is not specific to text or datasets. |\n| No-download cache inspection | [check_dataset_cache.py](scripts\u002Fcheck_dataset_cache.py) | Run when the user is offline, wants to know what is already cached, or needs signature\u002Fcache state without downloads or deletes. |\n| Safe text plotting smoke | [text_smoke.py](scripts\u002Ftext_smoke.py) | Run to confirm `yellowbrick.text` can render inline synthetic text plots with Matplotlib `Agg`; it uses no external data and no optional downloads. |\n\nExample safe runs from an environment where Yellowbrick is importable:\n\n```bash\npython skills\u002Fdisco\u002Fyellowbrick\u002Fsub-skills\u002Ftext-and-datasets\u002Fscripts\u002Fcheck_dataset_cache.py --dataset hobbies\npython skills\u002Fdisco\u002Fyellowbrick\u002Fsub-skills\u002Ftext-and-datasets\u002Fscripts\u002Ftext_smoke.py --outdir \u002Ftmp\u002Fyellowbrick-text-smoke\n```\n\n## Dataset quick map\n\n| Loader | Object | Shape or size | Best fit | Notes |\n|---|---|---:|---|---|\n| `load_bikeshare()` | `Dataset` | `X=(17379, 12)` | regression | Bike sharing demand examples. |\n| `load_concrete()` | `Dataset` | `X=(1030, 8)` | regression | Small, numeric, good for regressor demos. |\n| `load_credit()` | `Dataset` | `X=(30000, 23)` | binary classification \u002F clustering | Credit default data; can be large for quick CV examples. |\n| `load_energy()` | `Dataset` | `X=(768, 8)` | regression \u002F multi-output source data | Default `y` is the documented primary target. |\n| `load_game()` | `Dataset` | `X=(67557, 42)` | multiclass classification | Categorical features usually need encoding before sklearn estimators. |\n| `load_hobbies()` | `Corpus` | `448` docs, `5` labels | text analysis \u002F classification \u002F clustering | Returns a `Corpus`, not `X, y`; use `corpus.data` and `corpus.target`. |\n| `load_mushroom()` | `Dataset` | `X=(8123, 3)` | binary classification \u002F clustering | Categorical mushroom attributes need preprocessing for many estimators. |\n| `load_occupancy()` | `Dataset` | `X=(20560, 5)` | binary classification | Multivariate time-series-style occupancy data. |\n| `load_spam()` | `Dataset` | `X=(4600, 57)` | binary classification \u002F threshold analysis | Useful for threshold and imbalance demonstrations. |\n| `load_walking()` | `Dataset` | `X=(149332, 4)` | clustering \u002F time-series-style diagnostics | Large row count; sample for quick visual checks. |\n| `load_nfl()` | `Dataset` | `X=(494, 23)` | clustering | Football receiver clustering examples. |\n\n## Text visualizer quick map\n\n| Visualizer | Input required | Best fit | Constraints |\n|---|---|---|---|\n| `FreqDistVisualizer` | Count-vectorized document matrix plus feature names | Top-token counts in a corpus or by class | Use `FreqDistVisualizer`, not `FrequencyVisualizer`, in public guidance. |\n| `TSNEVisualizer` | Numeric vectorized corpus matrix | 2D document embedding and class\u002Fcluster separation | t-SNE is expensive; vectorize first and lower `perplexity` for tiny samples. |\n| `UMAPVisualizer` | Numeric vectorized corpus matrix | 2D document embedding with UMAP | Requires optional `umap-learn`; use `metric=\"cosine\"` for many text vectors. |\n| `DispersionPlot` | Tokenized documents: list\u002Fgenerator of token lists | Lexical dispersion of known terms across documents | Search terms must survive tokenization and casing. |\n| `WordCorrelationPlot` | Raw text documents: list\u002Fgenerator of strings | Word or phrase co-occurrence heatmap | Every requested word\u002Fphrase must appear in the corpus. |\n| `PosTagVisualizer` | Pre-tagged nested docs or raw text with parser | POS tag distributions | Prefer pre-tagged tuples for no-optional-dependency smoke checks. |\n\n## Canonical dataset pattern\n\n```python\nfrom yellowbrick.datasets import load_concrete, load_hobbies\n\n# Tabular data: pandas DataFrame\u002FSeries if pandas is installed, otherwise numpy arrays.\nX, y = load_concrete()\n\n# Request the object when you need metadata or explicit return control.\ndataset = load_concrete(return_dataset=True)\nX_np, y_np = dataset.to_numpy()\n\n# Text corpus loaders return Corpus objects, not X\u002Fy directly.\ncorpus = load_hobbies()\ndocs = corpus.data\ntarget = corpus.target\nlabels = corpus.labels\n```\n\n## Canonical text pattern\n\n```python\nimport matplotlib\nmatplotlib.use(\"Agg\")  # set before pyplot in headless sessions\n\nfrom sklearn.feature_extraction.text import CountVectorizer, TfidfVectorizer\nfrom yellowbrick.text import FreqDistVisualizer, TSNEVisualizer\n\ncount_vectorizer = CountVectorizer()\nX_counts = count_vectorizer.fit_transform(docs)\nfeatures = count_vectorizer.get_feature_names_out()\n\nfreq = FreqDistVisualizer(features=features)\nfreq.fit(X_counts, target)\nfreq.show(outpath=\"freqdist.png\", clear_figure=True)\n\nX_tfidf = TfidfVectorizer().fit_transform(docs)\ntsne = TSNEVisualizer(random_state=13, perplexity=10)\ntsne.fit(X_tfidf, target)\ntsne.show(outpath=\"tsne.png\", clear_figure=True)\n```\n\n## Safety checklist\n\n- Do not call a loader in offline mode unless its dataset directory already\n  exists in the selected cache; missing data triggers Yellowbrick's downloader.\n- Use the cache inspector before telling an offline user that a loader will work.\n- Use `return_dataset=True` only on tabular loaders; `load_hobbies()` already\n  returns a `Corpus` object.\n- Confirm whether `pandas` is installed before promising DataFrame\u002FSeries\n  returns; numpy arrays are the normal fallback.\n- Confirm text input state: raw strings for vectorizers and `WordCorrelationPlot`,\n  token lists for `DispersionPlot`, numeric vectors for TSNE\u002FUMAP, and nested\n  `(token, tag)` tuples for parser-free POS visualization.\n- Confirm optional dependencies before recommending `UMAPVisualizer` or raw-text\n  parser mode for `PosTagVisualizer`.\n- Save figures with a non-interactive backend and class visualizer API when the\n  output must be a file.\n","# Yellowbrick Text and Datasets\n\nUse this sub-skill when the task is about Yellowbrick dataset loading, dataset\ncache\u002Fdownload behavior, the `load_*` helpers, the hobbies corpus, or\ntext-specific visualizers. It covers `FreqDistVisualizer`, `TSNEVisualizer`,\n`UMAPVisualizer`, `DispersionPlot`, `WordCorrelationPlot`, and\n`PosTagVisualizer`.\n\nFor package-wide Yellowbrick lifecycle, axes reuse, style, and headless saving,\nread [shared visualizer patterns](..\u002F..\u002Freferences\u002Fvisualizer-patterns.md). For\nbroad install, Matplotlib backend\u002Ffont, and scikit-learn compatibility failures,\nread the root [troubleshooting reference](..\u002F..\u002Freferences\u002Ftroubleshooting.md)\nbefore applying text\u002Fdataset-specific fixes.\n\n## Use this sub-skill for\n\n- Choosing among Yellowbrick example datasets and matching them to regression,\n  classification, clustering, time-series-style, or text-analysis tasks.\n- Loading tabular datasets with `load_bikeshare()`, `load_concrete()`,\n  `load_credit()`, `load_energy()`, `load_game()`, `load_mushroom()`,\n  `load_occupancy()`, `load_spam()`, `load_walking()`, or `load_nfl()`.\n- Loading the hobbies corpus with `load_hobbies()` and using `Corpus.data`,\n  `Corpus.target`, `Corpus.labels`, or `Corpus.files`.\n- Deciding between default `X, y` returns, `return_dataset=True`, `to_numpy()`,\n  `to_pandas()`, and `to_dataframe()`.\n- Controlling or inspecting the dataset cache with `data_home`,\n  `YELLOWBRICK_DATA`, and the no-download cache inspector.\n- Using text visualizers after the corpus is already vectorized, tokenized, or\n  part-of-speech tagged.\n- Explaining optional extras: `pandas` for DataFrame returns, `umap-learn` for\n  `UMAPVisualizer`, and `nltk`\u002F`spacy` plus data\u002Fmodels for parser-based POS\n  tagging.\n\n## Route elsewhere when\n\n- The user has already loaded data and now wants classifier score plots such as\n  reports, confusion matrices, ROC\u002FPR curves, or threshold tuning: route to\n  [classifier visualizers](..\u002Fclassifier-visualizers\u002FSKILL.md).\n- The user wants residuals, prediction error, Cook's distance, or alpha\n  selection: route to [regressor visualizers](..\u002Fregressor-visualizers\u002FSKILL.md).\n- The user wants feature ranking, PCA\u002Fmanifold feature projections, class\n  balance, binning, or feature-target correlation: route to\n  [feature-target visualizers](..\u002Ffeature-target-visualizers\u002FSKILL.md).\n- The user wants clustering diagnostics, validation\u002Flearning curves, CV scores,\n  RFECV, feature importances, or dropping curves: route to\n  [cluster-model-selection](..\u002Fcluster-model-selection\u002FSKILL.md).\n- The user asks about contrib wrappers, decision boundaries, missing-value\n  contrib plots, or statsmodels adapters: route to\n  [contrib-and-extensions](..\u002Fcontrib-and-extensions\u002FSKILL.md).\n\n## Required read and run map\n\n| Need | Read or run | When to use it |\n|---|---|---|\n| Dataset names, task fit, `load_*`, return types, `Dataset`\u002F`Corpus`, cache rules, downloader flags, and offline behavior | [datasets reference](references\u002Fdatasets.md) | Read before giving loader code, cache advice, or no-network guidance. |\n| Text visualizer selection, exact input formats, optional extras, and safe file output | [text visualizers reference](references\u002Ftext-visualizers.md) | Read before using any `yellowbrick.text` visualizer or diagnosing text input-shape errors. |\n| Dataset cache\u002Fdownload\u002Fsignature\u002Foffline failures, missing optional dependencies, wrong class names, parser\u002Ftagset errors, and display\u002Ffont warnings | [text\u002Fdataset troubleshooting](references\u002Ftroubleshooting.md) | Read after any failure symptom or before advising a risky cleanup\u002Fdownload action. |\n| Common Yellowbrick `fit`\u002F`score`\u002F`transform`\u002F`show(outpath=...)`, Matplotlib `Agg`, axes, and style rules | [shared visualizer patterns](..\u002F..\u002Freferences\u002Fvisualizer-patterns.md) | Read before writing report-quality code or headless automation. |\n| Broad installation, backend, font, or scikit-learn compatibility issues | [root troubleshooting](..\u002F..\u002Freferences\u002Ftroubleshooting.md) | Read when the problem is not specific to text or datasets. |\n| No-download cache inspection | [check_dataset_cache.py](scripts\u002Fcheck_dataset_cache.py) | Run when the user is offline, wants to know what is already cached, or needs signature\u002Fcache state without downloads or deletes. |\n| Safe text plotting smoke | [text_smoke.py](scripts\u002Ftext_smoke.py) | Run to confirm `yellowbrick.text` can render inline synthetic text plots with Matplotlib `Agg`; it uses no external data and no optional downloads. |\n\nExample safe runs from an environment where Yellowbrick is importable:\n\n```bash\npython skills\u002Fdisco\u002Fyellowbrick\u002Fsub-skills\u002Ftext-and-datasets\u002Fscripts\u002Fcheck_dataset_cache.py --dataset hobbies\npython skills\u002Fdisco\u002Fyellowbrick\u002Fsub-skills\u002Ftext-and-datasets\u002Fscripts\u002Ftext_smoke.py --outdir \u002Ftmp\u002Fyellowbrick-text-smoke\n```\n\n## Dataset quick map\n\n| Loader | Object | Shape or size | Best fit | Notes |\n|---|---|---:|---|---|\n| `load_bikeshare()` | `Dataset` | `X=(17379, 12)` | regression | Bike sharing demand examples. |\n| `load_concrete()` | `Dataset` | `X=(1030, 8)` | regression | Small, numeric, good for regressor demos. |\n| `load_credit()` | `Dataset` | `X=(30000, 23)` | binary classification \u002F clustering | Credit default data; can be large for quick CV examples. |\n| `load_energy()` | `Dataset` | `X=(768, 8)` | regression \u002F multi-output source data | Default `y` is the documented primary target. |\n| `load_game()` | `Dataset` | `X=(67557, 42)` | multiclass classification | Categorical features usually need encoding before sklearn estimators. |\n| `load_hobbies()` | `Corpus` | `448` docs, `5` labels | text analysis \u002F classification \u002F clustering | Returns a `Corpus`, not `X, y`; use `corpus.data` and `corpus.target`. |\n| `load_mushroom()` | `Dataset` | `X=(8123, 3)` | binary classification \u002F clustering | Categorical mushroom attributes need preprocessing for many estimators. |\n| `load_occupancy()` | `Dataset` | `X=(20560, 5)` | binary classification | Multivariate time-series-style occupancy data. |\n| `load_spam()` | `Dataset` | `X=(4600, 57)` | binary classification \u002F threshold analysis | Useful for threshold and imbalance demonstrations. |\n| `load_walking()` | `Dataset` | `X=(149332, 4)` | clustering \u002F time-series-style diagnostics | Large row count; sample for quick visual checks. |\n| `load_nfl()` | `Dataset` | `X=(494, 23)` | clustering | Football receiver clustering examples. |\n\n## Text visualizer quick map\n\n| Visualizer | Input required | Best fit | Constraints |\n|---|---|---|---|\n| `FreqDistVisualizer` | Count-vectorized document matrix plus feature names | Top-token counts in a corpus or by class | Use `FreqDistVisualizer`, not `FrequencyVisualizer`, in public guidance. |\n| `TSNEVisualizer` | Numeric vectorized corpus matrix | 2D document embedding and class\u002Fcluster separation | t-SNE is expensive; vectorize first and lower `perplexity` for tiny samples. |\n| `UMAPVisualizer` | Numeric vectorized corpus matrix | 2D document embedding with UMAP | Requires optional `umap-learn`; use `metric=\"cosine\"` for many text vectors. |\n| `DispersionPlot` | Tokenized documents: list\u002Fgenerator of token lists | Lexical dispersion of known terms across documents | Search terms must survive tokenization and casing. |\n| `WordCorrelationPlot` | Raw text documents: list\u002Fgenerator of strings | Word or phrase co-occurrence heatmap | Every requested word\u002Fphrase must appear in the corpus. |\n| `PosTagVisualizer` | Pre-tagged nested docs or raw text with parser | POS tag distributions | Prefer pre-tagged tuples for no-optional-dependency smoke checks. |\n\n## Canonical dataset pattern\n\n```python\nfrom yellowbrick.datasets import load_concrete, load_hobbies\n\n# Tabular data: pandas DataFrame\u002FSeries if pandas is installed, otherwise numpy arrays.\nX, y = load_concrete()\n\n# Request the object when you need metadata or explicit return control.\ndataset = load_concrete(return_dataset=True)\nX_np, y_np = dataset.to_numpy()\n\n# Text corpus loaders return Corpus objects, not X\u002Fy directly.\ncorpus = load_hobbies()\ndocs = corpus.data\ntarget = corpus.target\nlabels = corpus.labels\n```\n\n## Canonical text pattern\n\n```python\nimport matplotlib\nmatplotlib.use(\"Agg\")  # set before pyplot in headless sessions\n\nfrom sklearn.feature_extraction.text import CountVectorizer, TfidfVectorizer\nfrom yellowbrick.text import FreqDistVisualizer, TSNEVisualizer\n\ncount_vectorizer = CountVectorizer()\nX_counts = count_vectorizer.fit_transform(docs)\nfeatures = count_vectorizer.get_feature_names_out()\n\nfreq = FreqDistVisualizer(features=features)\nfreq.fit(X_counts, target)\nfreq.show(outpath=\"freqdist.png\", clear_figure=True)\n\nX_tfidf = TfidfVectorizer().fit_transform(docs)\ntsne = TSNEVisualizer(random_state=13, perplexity=10)\ntsne.fit(X_tfidf, target)\ntsne.show(outpath=\"tsne.png\", clear_figure=True)\n```\n\n## Safety checklist\n\n- Do not call a loader in offline mode unless its dataset directory already\n  exists in the selected cache; missing data triggers Yellowbrick's downloader.\n- Use the cache inspector before telling an offline user that a loader will work.\n- Use `return_dataset=True` only on tabular loaders; `load_hobbies()` already\n  returns a `Corpus` object.\n- Confirm whether `pandas` is installed before promising DataFrame\u002FSeries\n  returns; numpy arrays are the normal fallback.\n- Confirm text input state: raw strings for vectorizers and `WordCorrelationPlot`,\n  token lists for `DispersionPlot`, numeric vectors for TSNE\u002FUMAP, and nested\n  `(token, tag)` tuples for parser-free POS visualization.\n- Confirm optional dependencies before recommending `UMAPVisualizer` or raw-text\n  parser mode for `PosTagVisualizer`.\n- Save figures with a non-interactive backend and class visualizer API when the\n  output must be a file.\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\u002Fyellowbrick\u002Fsub-skills\u002Ftext-and-datasets",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 text-and-datasets",{"tool":40,"command":41},"bunx","bunx skills add VectorSpaceLab\u002FAREX-Skill --skill text-and-datasets",{"tool":43,"command":44},"pnpm","pnpm dlx skills add VectorSpaceLab\u002FAREX-Skill --skill text-and-datasets","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fyellowbrick\u002Fsub-skills\u002Ftext-and-datasets\u002FSKILL.md",[47,53,57,61,67],{"path":48,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":51,"sourceHtmlUrl":52},"references\u002Fdatasets.md","REFERENCE","参考文档",9872,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fyellowbrick\u002Fsub-skills\u002Ftext-and-datasets\u002Freferences\u002Fdatasets.md",{"path":54,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":55,"sourceHtmlUrl":56},"references\u002Ftext-visualizers.md",10547,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fyellowbrick\u002Fsub-skills\u002Ftext-and-datasets\u002Freferences\u002Ftext-visualizers.md",{"path":58,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":59,"sourceHtmlUrl":60},"references\u002Ftroubleshooting.md",11288,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fyellowbrick\u002Fsub-skills\u002Ftext-and-datasets\u002Freferences\u002Ftroubleshooting.md",{"path":62,"resourceType":63,"resourceTypeDesc":64,"sizeBytes":65,"sourceHtmlUrl":66},"scripts\u002Fcheck_dataset_cache.py","SCRIPT","脚本",8756,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fyellowbrick\u002Fsub-skills\u002Ftext-and-datasets\u002Fscripts\u002Fcheck_dataset_cache.py",{"path":68,"resourceType":63,"resourceTypeDesc":64,"sizeBytes":69,"sourceHtmlUrl":70},"scripts\u002Ftext_smoke.py",6373,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fyellowbrick\u002Fsub-skills\u002Ftext-and-datasets\u002Fscripts\u002Ftext_smoke.py",[72,79,86,93,100,107],{"skillNo":73,"name":74,"nameZh":6,"description":75,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":76,"categories":77,"keywords":78,"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":80,"name":81,"nameZh":6,"description":82,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":83,"categories":84,"keywords":85,"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":87,"name":88,"nameZh":6,"description":89,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":90,"categories":91,"keywords":92,"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":94,"name":95,"nameZh":6,"description":96,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":97,"categories":98,"keywords":99,"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":101,"name":102,"nameZh":6,"description":103,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":104,"categories":105,"keywords":106,"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":108,"name":109,"nameZh":6,"description":110,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":111,"categories":112,"keywords":113,"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",[],[]]