AGENT SKILL

documents-and-visualization

Work with spaCy Doc, Token, Span, DocBin, tokenization, matchers, serialization, scoring, and displaCy visualization.

License 许可
MIT
最近更新
2026-09-03
Stars
★ 351
Forks
27

01 前言 Front Matter

name
documents-and-visualization
license
MIT
metadata
{ "disco-role": "operating" }
description
Work with spaCy Doc, Token, Span, DocBin, tokenization, matchers, serialization, scoring, and displaCy visualization.
disable-model-invocation
true

02 资源文件 7

03 SKILL.md 正文

documents-and-visualization

Use this sub-skill for the core spaCy document model: tokenization, Doc/Token/Span APIs, DocBin serialization, rule-based matching, EntityRuler/SpanRuler, scoring, and displaCy HTML rendering.

Read first

What this sub-skill covers

  • spacy.blank() and spacy.load() for processing text with an existing package or local pipeline.
  • nlp(text) and nlp.pipe() for document streaming and as_tuples context.
  • Doc, Token, Span, extension attributes, char_span, retokenize, and DocBin round-trips.
  • Tokenizer behavior, language data, and blank-language caveats.
  • Rule-based matching and entity/span rulers for token-pattern workflows.
  • displaCy rendering for entities, dependencies, and spans.

What to route elsewhere

  • Install/import, version, CLI entry-point, missing model package, or backend questions: install-and-inspect.
  • Custom pipeline factories/components, add_pipe, registry wiring, or pipe analysis: pipeline-components.
  • init config, debug config, convert, train, evaluate, package, or validate: training-and-cli.
  • spacy project clone/assets/run/push/pull/DVC workflows: project-workflows.

Quick use pattern

  1. Start from the object you need: Doc, Span, Token, matcher, ruler, or displacy.
  2. Check the exact signature in the reference before assuming optional parameters.
  3. Run the bundled smoke script if you need a safe no-download proof of behavior.
  4. Use the troubleshooting reference when offsets, tokenization, or serialized artifacts do not line up.

Evidence basis

This sub-skill is distilled from spaCy's public docs, verified installed-package signatures, core token/matcher/serialization source, and behavior-backed tests. It is self-contained and does not require reopening the source checkout.

04 同仓库更多 Skills

VectorSpaceLab
VectorSpaceLab/AREX-Skill

passnet-triton-opt

Make a single PassNet Triton kernel fast AND numerically faithful: performance model (when a replacement can win at all), block/grid/warp 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.

★ 351
VectorSpaceLab
VectorSpaceLab/AREX-Skill

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.

★ 351
VectorSpaceLab
VectorSpaceLab/AREX-Skill

passnet-skill

PassNet GPU kernel optimization via compiler passes. Design and implement Triton-based optimization passes, create pass files under ./pass_dir/, self-evaluate with pass_evaluator, and iterate to maximize GPU speedup.

★ 351
VectorSpaceLab
VectorSpaceLab/AREX-Skill

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()/replacement_args()/replacement_func() files, when a pass "failed to match", or when deciding how to fuse multiple kernels/ops.

★ 351
VectorSpaceLab
VectorSpaceLab/AREX-Skill

passnet-orchestrate

Round/batch planning and decision support for PassNet graph optimization. Use for multi-sample triage, worker allocation, eval-budget policy, and post-eval keep/revert 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.

★ 351
VectorSpaceLab
VectorSpaceLab/AREX-Skill

passnet-feedback

Fast, deterministic feedback for PassNet work: pre-flight pattern/match 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).

★ 351