missingno
Guides missingno missing-data visualization, nullity filtering and sorting, package smoke checks, and troubleshooting for pandas DataFrames.
- License 许可
- MIT
- 最近更新
- 2026-09-03
- Stars
- ★ 351
- Forks
- 27
01 前言 Front Matter
- name
- missingno
- license
- MIT
- metadata
- { "disco-role": "operating" }
- description
- Guides missingno missing-data visualization, nullity filtering and sorting, package smoke checks, and troubleshooting for pandas DataFrames.
- disable-model-invocation
- true
02 资源文件 5
- 参考文档references/maintainer-notes.md3.0 KB
- 参考文档references/repo-provenance.md1.9 KB
- 参考文档references/repo-routing-metadata.json399 B
- 参考文档references/troubleshooting.md3.9 KB
- 脚本scripts/missingno_smoke_check.py4.9 KB
03 SKILL.md 正文
missingno Repo Skill
Use this skill when a task involves the Python package missingno, missing-data
visualization, nullity plots, or DataFrame completeness filtering/sorting. It
is an operating guide for future Researcher sessions; it is not a maintainer
checkout dependency.
First checks
For ordinary use, install the public package and verify imports:
python -m pip install missingno
python - <<'PY'
import missingno as msno
print(msno.__version__)
print([name for name in ["matrix", "bar", "heatmap", "dendrogram", "nullity_filter", "nullity_sort"] if hasattr(msno, name)])
PY
Expected public API in this snapshot:
msno.matrix(df, ...)msno.bar(df, ...)msno.heatmap(df, ...)msno.dendrogram(df, ...)msno.nullity_filter(df, ...)msno.nullity_sort(df, ...)
There is no verified package CLI entry point and no verified geoplot export in
this repository snapshot.
Route map
| User task | Read next |
|---|---|
| Choose or configure a nullity matrix, bar chart, heatmap, or dendrogram | sub-skills/visualizations/SKILL.md |
| Interpret nullity correlations, dendrogram clusters, time-index frequency ticks, labels, sparklines, or headless plotting | sub-skills/visualizations/SKILL.md |
| Filter to the most/least complete columns or sort rows/columns by completeness | sub-skills/nullity-utilities/SKILL.md |
Understand filter, n, p, sort, or axis behavior used by plot APIs |
sub-skills/nullity-utilities/SKILL.md |
| Diagnose installation, import, dependency, headless plotting, no-CLI, stale-docs, or smoke-check failures | references/troubleshooting.md |
| Edit or test the upstream repository rather than just use the package | references/maintainer-notes.md |
| Decide whether this generated skill is stale for a checkout | references/repo-provenance.md |
Shared smoke helper
Run the bundled smoke helper when an environment should be checked without network downloads or a notebook display:
python scripts/missingno_smoke_check.py --skip-plots
MPLBACKEND=Agg python scripts/missingno_smoke_check.py --plot all --output-dir /tmp/missingno-smoke
The helper uses a synthetic pandas DataFrame, validates nullity_filter and
nullity_sort, then optionally renders matrix/bar/heatmap/dendrogram plots via
a headless matplotlib backend. Read the helper before adapting it:
scripts/missingno_smoke_check.py.
Operating boundaries
- Prefer small synthetic DataFrames for examples and diagnostics. The public README's collision dataset example is useful conceptually but uses network data and is not required for package operation.
- Keep
SKILL.mdfiles as routers. Use the linked references for API tables, examples, and troubleshooting details. - Do not claim
inline=orgeoplotsupport for this snapshot: installed signatures and exports did not verify those capabilities even though older documentation text mentions them. - Plotting APIs return matplotlib
Axesobjects. In scripts or CI, setMPLBACKEND=Agg, save figures explicitly, and close figures after checks.
Public provenance and router metadata
- references/repo-provenance.md records the source commit/tag, package version, dirty-state summary, and evidence paths.
- references/repo-routing-metadata.json contains structured metadata for a future managed repo-skill import. This run intentionally does not import the skill.
04 同仓库更多 Skills
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.
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.
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.
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.
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.
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).